Singapore Food Guide: 25 Must-Eat Dishes (& Where to Try Them)


First one would be Old Airport Road hawker center, I like won ton mee here very much and you can also try some other dishes in your list here as well. The food served is exceptionally delicious.

Bersama Kami

Both areas are good for tourist sight seeing as well!! Tue to Thur 12pm — Hill Street Char Kway Teow. Closed on Mondays and Thursdays. Do go for their pork satay, but be prepared to wait for the satays. I prefer this stall to the other stalls at Chomp Chomp, but there are quit a number of stingray stalls there. Hope you guys enjoy your stay in Singapore Mark! Hi Mark, Here are a list of recommendations. Still the best chicken imo, rice and chilli are also a good standard.

I know you like chilli, famous for its great noodles and bright orange napalm. Great duck, char siu and roast pork. Eaten by presidents and kings. Go early as they do sell out. Geylang Serai Market is seriously underrated. These 2 stalls are Malay institutions. Old school roti prata! I know what some Singaporeans are thinking.. Bottom line is its very good and its true blue peranakan. East Coast Lagoon Food Village is an out of the way but another seriously underrated hawker centre.

Carrot Cake- Lagoon Carrot Cake. Best Chili crab I have ever eaten beats Jumbo imo as it is not so sweet and great sambal on their stingray. PS — they are not cheap but definably cheaper then jumbo and other big seafood chains. Do let us know if you will have a meet up, think that would be lots of fun. Hi Mark,If you are in Singapore,You must not miss eating mee rebus,mee siam,lontong,gado-gado,tosai. If you want eat the best tosai you should go to serangoon road little india,lontong,gado-gado,mee siam and mee rebus you should go to geylang food centre market at geylang serai.

Try their Deer Murtabak and Mutton Murtabak. You can request for double which means more meat. For super duper power delicious nasi padang you must go to Warong Pariaman located at North Bridge Road,try their grilled chicken cooked in coconut milk,bagedil,sotong sambal and ayam kalio.

Its better for you to come before Do come early around 5pm to avoid night crowds. For romantic grilled seafoods such as grilled sambal stingray you must go to East Coast Park Food Centre. Windy breeze and goes well with coconut juice. It will be worth your trip to all the places i mention and recommend. It is the epitome of a tourist trap; it has sub-par quality food with infamy for exorbitant prices charged to unsuspecting travellers.

Only advising against Newton in specific. For visiting FCs as a holistic experience and not just homing in on one single good stall in it, however, it is also not a bad option. They offer good quality food to a stunningly stupendous degree of variety. Picture chowing on sambal stingray at the side, while munching on glistening grilled chicken wings, not forgetting sharing a plate of chai tow kway, shredded chicken porridge, assorted meat satays with peanut sauce and onions and cucumbers, and of course washing that all down with an opened coconut.

These things are soo aromatic I could die. They can be smelled from afar as they boil in the hot dark soup. They are these brown eggs slow cooked to perfection. But if you were to encounter them, do try them.

Really happy to know that you are visiting Singapore. I promise you its the best dish ever!!! Also, try Laksa in Katong its a place and everyone knows about it. Signature Laksa and the best! They are very famous and the best part of the dish is the fragrant harmony of chilli and chinese black vinegar. Its located pretty near to the city so it should be fairly accessible for you.

Hi Mark, seems like i know you already after watching your vlog. It seems to me you had a huge fan base in Singapore, so needless to recommend you more places to eat. Hope you enjoty your feast in Singapore. Your post is amazing! Thanks for sharing this food list, because Singapore is my favorite and i like durian, but i have no idea about other varieties. This list is very helpful for those people who like to eat different varieties food of Singapore.

It is the best hokkien mee without doubt! Its a teocheow-style noodle dish. Flat noodles tossed in lard, vinegar and chilli, topped up with fishballs, fish cake and some pork.

They only open in the mornings from Mon-Sat. They are sold out by 11am. Please do yourself a favour by trying that. You can go to newton food centre near newton mrt station. You can find sambal stingray , satay , chilli crab and all different local food there!

You got to visit Fengshan too! For all the bbq fish, oyster omelette and bak chor mee Mince pork noodle you can try! As for Laksa, sure by all means try the Katong Laksa one. I like their broth which is really rich because of loads of coconut milk but if you do have time, do pay a trip to Sungei Road Laksa as well, which is more traditional and they do everything on a charcoal stove! For crab; I think chilli crab is a bit too sweet for me and thus I will really recommend black pepper crab at Eng Seng Restaurant as recommended by others.

Fish head curry here is a must, located on Dempsey Road, and the bonus is that the lime juice is soo good. Personal favorite is the one next to Rex cinema at Mackenzie Road. Its a small food court. The popular ones you should go is Katong Laksa. They have a few outlets around SG. Boon lay power nasi lemak! The queue can get very long so go early. They wont open until early evening. Located at b Boon Lay Place. Something similar to Nasi padang.

Its called Nasi Ambeng. Rice serves with various dishes in a big tray and can be shared with 2 or more people. Its the hype lately here. You should give it a try at Hajjah Mariam restaurant located at Westgate Basement level 2. Nearest mrt line would be Jurong East! U really got to try this fews dish of crab U will nv regret it for the trip here.

It will b a long Q u got to go super early to get a sit if not got to wait for hours. Mellben Seafood has made its way from a humble stall along Alexandra Road to a largely well-known seafood place with three outlets in Singapore. Its menu features a wide selection of crab dishes such as chilli crab, black pepper crab, claypot crab vermicelli soup, butter crab, salted egg yolk crab and shimmering sand crab.

Chilli crab at Mellben is one of the best in our Lion city. You can always get very big crabs in thick, spicy and slightly sweet chilli sauce. Just request for a less spicy plate if you cannot bear high level of spiciness.

Best place to have claypot crab bee hoon: Claypot crab bee hoon also known as claypot crab vermicelli soup is basically a noodles soup dish, highlighted by milky broth, which is rich with the aroma and sweetness from crabs.

For many locals, Mellben is the best place to enjoy claypot crab bee hoon — a crab dish that may be new to your culinary dictionary but can make you melt from the first spoon of soup. Close your eyes and imagine a big crab coated with thick, buttery, sweet and slightly spicy sauce.

It is a dish to die for. Ambience Mellben Seafood at Ang Mo Kio is far away from the bustling centre of Singapore, but the atmosphere herein is never quiet. Situated at a kopitiam in one of the HDB estates and adorned with decorated crab shells on the walls, Mellben provides an intimate dining ambience for crab lovers.

Notes — There are no nearby MRT stations so public transport accessibility is by bus — It is recommended checking prices of all the dishes before ordering as they are not listed on menu — It is always worth calling ahead to make a booking — Be prepared that the service of the store is almost at food court standard.

You have to try Hill Street Tai Hwa pork noodle mee pok dry. Also, you HAVE to try this minced meat noodle soup at bedok. There is usually a really long queue though!

Xing Ji rou cuo mian Address: First one would be Old Airport Road hawker center, I like won ton mee here very much and you can also try some other dishes in your list here as well. I wish you to enjoy Singaporean food, I like your videoes and looking forward to see more.

Sungei Road Laksa Address: Marsiling Lane, Block 19 , Singapore Hours: Hillstreet Char Kway Teow Address: Hai Sheng Carrot Cake Address: Kok Kee Wanton Mee Address: Tai Hwa Pork Noodle Address: Block Crawford Lane Singapore Hours: Simon Road Oyster Omelette Address: Guan Hoe Soon Restaurant Address: Block Commonwealth Crescent Singapore Hours: Rumah Makan Minang Address: Selera Rasa Nasi Lemak Address: Long queue and huge take-aways order.

The Roti Prata House Address: Please call to make reservation in advance. May have to wait up to 1. Ocean Curry Fish Head Address: I would suggest you to try Nasi Briyani, I watched a video of yours having kao mok gai near silom soi You must definitely try the traditional Chinese pastries from tong Heng confectionery shop. When in Singa — satay and chilli crab are a must for us, but all your list look mouth wateringly delcous.

Nowadays, generic prata shops use instant doughs for convenience, however this couple still makes their own dough.. Prata is a staple breakfast Singaporean dish anyway. If you are looking for Prata, Murtabak, Thosai and other Indian food in one comfortable place. The Pulled tea is good as well!

Very delicous Bak Kut Teh: Chicken Rice — No recommendaton 5. Wan Ton Mee — No recommendaton 8. Fish Bee Hoon — No recommendaton 9. Bak Chor Mee — No recommendaton Yong Tao Foo — No recommendaton Roast Duck — No recommendaton Peranakan Food — No recommendaton Zi Char Meal — No recommendaton Ice Kachang — No recommendaton Durian — Many stall along Geylang Road Nasi Padang — No recommendaton Nasi Lemak — No recommendaton Chili Crab — Roland Restaurant Sambal Stingray — No recommendaton Fish Head Curry — No recommendaton.

Hey mark Great to hear you coming to SG Seems You got lots of fans who wanna do a meet up Your welcome to use my cafe space Let me know. Hoe Nam Prawn Noodles: Laksa I personally loved the Katong Laksa located in Queensway Shopping Centre and you definitely wanna go with a Otah to go along with it.

And you can grab a bus down to Chinatown which have a market called Maxwell market over there there is satay, rojak and many local tasty food to end your dinner there. Make sure to get fried buns to mop up the gravy. Got one singapore lor mee very nice, only open at morning. The place at old airport road 51, hawker centre. Laksa Not so much a recommendation as a cautionary note: It simply lacks the flavour that umami-ness it used to have—even after adding lots of chili paste.

Meat here is fatty, tender, and flavourful. Hokkien Mee Bedok Block The old man who mans the cash register is Hainanese! The meat here can taste a bit dry, so be sure to add ginger and chili. There is a Cantonese style of chicken rice, easily identified by the juicier meat that has been blanched, not cooked all the way through, and liberally seasoned with light soy sauce. Tastier than the original style because in Singapore the quality of chicken meat is not exceptional.

Here, they mill the rice themselves to make carrot cake from scratch! After the traditional taste at Yap Kee, head over to RedRing Wanton Mee at 46 Holland Drive—this stall uses more modern technology like a ramen cooker to cycle the water out and avoid the alkaline flavour that is present in most regular wanton mee. Still, both stalls were fantastic and I used to eat from them nearly every day when I was working in the area. This is a modern take on Peranakan food by a young chef who is, of course, a Peranakan himself.

Aside from the many different curries, do try and make space for the tumeric chicken wings and buah keluak ice cream. Be not fooled by the other nasi lemak stall right next to this one!

The crispy style is more appealing to Chinese, who enjoy prata with utensils; the chewy kind is what the Indians prefer. Originally prata was eaten with bare hands, and you should be able to drown the whole prata in curry, then tear off chunks with a single hand.

Sin Ming is more of the latter camp. These following dishes are probably better eaten outside of Singapore: A good kopi seems particularly elusive nowadays. All the above are some pretty good local eats that I find are uncommon recommendations to visitors of Singapore, and I think the other commentors would not often give as recommendations.

They are all delish, so with the 25 dishes you already have, perhaps you can look at the above list I made for have an adventurous dish or two from this exotic list. Hey mark, if u like salted baked chicken. U can try Marina Square foodcourt, ah lam specialist! They are specialist in abalone noodles and salted baked chicken rice! Have fun in Singapore! For chicken rice, should head to maxwell hawker centre at stall Tian Tian Hainanese chicken rice.

Delicous Fish bee hoon also can find at maxwell too. For kaya toast, should eat it at Ah Kun Kaya toast, which is a popular and old brand that has many outlets around Singapore. If want something extra, Hong Lim hawker centre has one stall serve delicous curry chicken noodle. Hi Mark, you can try the Indian rojak at Tampines, its called Mahboob. Its opposite Tampines Polyclinic but be prepared for a long wait.

Once I waited for 90 mins during the fasting month. Hi Mark, the Best chicken rice is at golden mile food center, Hainanese Boneless chicken rice. Heres a link for your reference, https: You can go try special shanghai dim sum at maxwell food centre. The fried and steam dumplings as well as the spicy sour soup is awesome. A Must Try vvvvvv nice u will nv regret it. We waited a good 35 minutes for this on a Saturday night, but it was so worth the wait!

This bowl of meesua came with a generous amount of pig organ and lean meat. The soup is thick and flavorful with a hint of herbal taste but not too heavy at all. I wiped out the whole bowl! Try the best of the best dish according to how it should be, Chicken rice has 2 Major types, 1 Steamed White, 2 Roasted. Steam White Chicken Rice: Its the only thing i would eat if its my last dish on earth!

I would advise you to go early in the morning to avoid the lunch crowd, and it is closed on Monday! Both of the restaurants are located on Purvis Street. The kway chap in sg is pretty different from the one in bangkok but both are really delicious.

The portions here are big enough and defintely good quality. Absolutely worth a try! Peranakan food in Singgapore is no longer authentic. You should also go the Kovan Food Centre. I recommend both to you because of their spicy Chili and knowing what a chilli fanatic you personally are! Keep up the good work Mark! Some of my suggestions and my personal favorites: It is truly uniquely Singapore where all the local and overseas dishes are consolidated into one centre for the local to eat.

The roast duck amongst other dishes is fantastic! I recommend that you go for lunch there as the duck will usually run out by dinner due to reservations etc. Excited to hear that you both are making your way down to SG. Ok so the places to try out…. Portion is good for one person. Sold by this granny and grandpa. Old school fruit rojak. They sells this other peanut pancake-ish dessert that is good too. All 3 naan shops in Tekka are good.

You can follow the long queue or just go to the lesser ones. Avoid office lunch hour between 12 to 1pm, you can hardly find seats around those time. If you walk further down near the Little India Arts centre…there is an old school chapati place there. The shop is adjacent of the building. The teh tarik and ice milo are very sweet though. You can ask if they have sambal belacan, usually it is free. That chapati and that keema curry are so delicious. Plus, my late mother loved their chapatis which are always freshly made and soft to chew.

I would recommend Briyani Express in geylang serai too but i say forget it cos it sells out damn fast by 12 noon. Plus, briyani in tekka market are the best!

Actually, you cannot go wrong with food stalls in Tekka Market and Geylang Serai for any halal food. Yes, my friends and I tried almost all the stalls there during our lunch, dinner breaks and off days. Just opposite of Geylang Serai Market. Closed on Mondays i think. Their signature dish is snails in yellow curry. Sambal belacan and fresh malay style salad is free. And you can try the kuehs and the sweet bubur hitam there too. Go to the 2nd lane of the food market….

Or you can try the very over-rated and pricey Apollo banana leaf restaurant along race course road which is near the arts centre. The masala thosai there is good but the thali rice set, not that fantastic. The briyani is really a lot for one person…like a lot!! I guess Shakuntala is just very generous with their portions. Ayam buah keluak dish reminds me of our Javanese Kuah Rawon only thicker.

So this is the next better one. The seafood dishes and all others like oysters omelette etc can be found there. After eating, you can walk along the beach and unwind or vice versa. Someone mentions not to follow the food hype here in Singapore.

That person is right. And you can skip dining in food courts. Expensive and quite bland food. Eat there cos the bakmie dishes are delicious! Authentic Indonesian street food. Hey Hidayah, this is an awesome list of food suggestions, thank you very much for taking the time to write these amazing recommendations. Will try to visit a few of them.

I watch your videos all the time!! I love your videos. From the list youre missing something Mark, dont miss bonesteak-beach road please. Whenever i eat the bone steak, id go, if mark comes here he has to try this!

And putu piring-geylang serai, this ying will love. Id love for you to have a meet up session. Theres so much food for you to indulge in, in singapore. Hi Mark, Instead of recommending you what to eat best only.

I would also share with you the places i go for certain food types. It certainly worth visiting and experiencing. Its also a place where you could try lots of different malay food under one roof.

My fav would be. Then upon order, they pour the coconut gravy over the chicken. Just try the mutton briyani. I always buy both cos the sauce is awesome. Also here sells kentang ball which is a giant potato stuffs with beef in broth. Cant find it elsewhere. They have been in business for more than 15 years. Coconut rice with side dishes such as chicken wing and otah otah bbq minced fish.

I would recommend the red version. You could even ask for mix kway teow and bee hoon rice vermicelli. Or if you wanna try something different can go for Maggi goreng.

Here they are famous for their special combo. Haron 30 Satay Chicken Wing: You gotta try their chicken wing. It goes very well with their mutton satay. I always order mutton satay and avoid chicken. Telling them to bbq it slightly crispier. This eaten with rice is the so awesome. Especially by the beach for dinner. Squeeze lime , and dip in cincalok fermented shrimp dip.

Norris Road — Azmi Restaurant. Best place for breakfast. Just look at the staff working there. You can tell its gonna be authentic taste for each visit. Tekka Market and Food centre: Their sauce is Awesome. Its quite similar to Penang Pasembur forgive me if my spelling is wrong. Instead of the sauce being lathered all over your choices. Ours is you dip it before you eat.

Lastly do enjoy yourself and have a great time in SG. I taken uni in Singapore for 4 years and my numb 1 place to eat is hotpot buffe in near bugis station. I would love to see you eat there and record a blog about it. Believe me, this is the must try pot…. Specaily beef and lamb for the spicy hot soup Here is the information about the place: Christian have asked me to introduce you the best of singapore! Ananda Bhavan is an Indian Vegetarian Restaurant.

The rice is fragrant and carries an explosion of spices. Their side dishes for the biroyani varies from day to day, from vadai, to long beans, to veggie balls and more. All the spicy goodness you need. It also comes with 2 small bowls of curry and dhal, 1 small bowl of Raita yogurt with onions, carrots etc and also 1 small bowl of dessert to top it all off. A quite place to come back to at night to escape from the hustle and bustle of the city.

For desserts, there is a famous dessert stall called Qing Tian at Redhill hawker centre, just about 5 min walk from Redhill mrt station. My daddy and I really enjoy watching your vlogs it has become a routine night bonding activity for us now , and we really like how genuine and descriptive u are and my dad looooves watching you eat, he says it makes him so satisfied just watching you INDULGE and DROWN in all the flavours of your food, so keep up the awesome work and I hope to see the same enjoyment and indulgence when you try singapore food!

The prawns are fresh and soup is so so good. Have fun in SG! Mark beware of Fried Singaporean Noodles with vanilla ice cream!! The best malay style chicken rice got to be Mat Noh n rose whampoa drive food center.. Garlic chili or spicy soy sauce..

Tian Tian chicken rice at maxwell market is famous too! Its a plate of chicken rice i have nver taste before else where. If you are eating there, drop me an email and I will try my best to meet you!

Oh and have a good one! They also do a seriously good Hokkien Mee. Ah roi Mark Mark Khrub.. Try to eat at kamasutra, loacted nearby to Marina bay sands.. And try eating chicken rice at boon tong kee, they had several placed in SG, they had some really delicious menu.. And try to eat nasi lemak at holland village.. Tryout the indonesia foods here too make like nasi rawon, nasi ambeng etc at the famous geylang markets. They are in the same row of atalls in the same market. In short u should go to the geylang market and tekka market… 2 places will rows of local food and dessert…unlimited choices of malay and indian food there and some are awardwinning stalls…the best food is in the local hawker stalls in these markets.

They also make the chili crabs. Delicious Boneless Chicken Rice Address: Carrot Cake Chai Tow Kway: Blk 85 Bedok North Street 4 , Singapore Have a great trip! By the way, if you are planning to go for the chilli crab, please do not forget that you could also make an order of the black pepper crab as well!!! Although you have just published a vlog about the Thai Black pepper crab! Swee Choon Dim Sum. Love your list but you are missing my favorite dish in all of Singapore.

Now I know you listed chili crab in your top 25 but you must try the black pepper crab. The black pepper sauce is so delicious it goes so well with the fried rice. Now the restaurant opens from 4: Why just settle for a normal chicken rice?

Try a special chicken rice that is cold. This is my favorite prawn noodles restaurant ever, i used to go there when my worked part time in the area and i eat lunch, sometimes, going cycling at east coast park which is a bus ride away. The lines are usually at Tian Tian stall, but if you do a bit of research online, you will find that the original chef from Tian Tian had a dispute with the owner and left to start his own stall….

Anthony Bourdain went to Ah Tian during his visit to Singapore and absolutely loved it. I personally have made a second trip to Singapore all the way from San Francisco! JUST to get the chicken rice again. Perhaps it would be an interesting experiment to try both side to side in your video, as the two operations are only separated by a few stands. I have tried both and they are definitely top 2 in Singapore. I am currently in Taiwan and have been referring to your Taipei video series as a guide, thank you for your insight and research!

Hi Mark, Based in Denmark now, but spent half my childhood in Singapore. At Blk , there was this fantastic Nasi Padang stall in the coffee shop, named after the founder , Che which means Madam in Malay Rose. Was back in Singapore in Nov , and found out that they had moved to Block Her daughter runs the stall now. Be sure to go before 1 pm as they sell out quickly, and check when they have their day off.

Mdm Rose migrated to Singapore as a child with her parents from Sumatra in the s, and her parents started selling Nasi Padang back then.

She Is one of several siblings who started Nasi Padang stalls. I believe that her sisters run a more famous Nasi Padang joint in Kampong Glam. Daily 10am — 10pm Web Link: The laksa is so good that their story and history are also mentioned in heritage tours for schools.

Their laksa is known for having their noodles cut, served with thick laksa gravy, allowing it to be scooped with only a spoon! Don't forget to add a piece of Otah! If the stall is too out of the way, they have multiple outlets over SG, check out their website to ensure that you do not patrionise a fake stall claiming to be the famous Katong Laksa!

Yes they are so famous that stalls all over SG claim to be from them. Many might think that SongFa has been overhyped and commecialised due to expanding outlets.

However, in my opinion SongFa has been consistent with it's taste and flavor, and is easily accessible with around 4 outlets in SG. I recommend this outlet as the first outlet, and it is often the most busy and popular. Also add the salted vegetables for a perfect combi! Don't forget that the soup is refillable! You might even want to have a walk over at Clarke Quay at night just across the street:. As I know you love spicy food Mark, this stall's Hokkien Mee uses much more spicy chilli, which is different with the conventional Hokkien Mee chilli which is more towards the sweet side.

Their chilli enhances the noodles to a whole new level and gives you a very exciting and delicious taste experience.

Every time Chicken Rice is mentioned, Tian Tian never fails to top the charts. Ask any Singaporean and most will agree that Tian Tian has got to be one of the best. The trip down here will definitely be worth it! Do look out for other famous stalls here! I'll stop here for now Mark, will be adding on more recommendations tommorrow or the day after:.

Mee Rebus, Lontong, mutton bones are some other things worth considering. Hope Singapore readers can suggest some places where these are good. C est excellent voir exquis!! See you later from didier france. U should check Tis coffee shop out they r famous for: Ask for extra spicy!! I bet u will love it. Lastly do leave some space for durian down e road. Head down u wil pass by a down ahead u will see e durian store called durian collection!

Be sure to ask for bitter durian!!! They r famous for ripping tourist!! As for tradition Kaya toast- Ya Kun is definitely e best choice! U can grab it once u reach our changi airport too! Hope u n Ying will haf a great experience here in Singapore. Also really hope we can haf a meet up session too!! Welcome to Singapore Mark and Ying! Blk 16, Bedok South Road, , Singapore http: Therefore, I personally can attest to the consistent quality and delicious taste from these places.

At least to me, this is possibly the best sambal stingray I ever tried in Singapore have been patronising the stall for over 10 years. Both their sambal and chincalok dipping sauce are homemade! Their sambal has a fiery kick and fragrant note to it while the chincalok has a harmonious balance of spiciness and sourness.

Do order their stir fried kangkong or potato leaves too. Chan BBQ- Bedok Fengshan Market Block 85 At this market, you can at the same time try the minced meat noodle soup version that the market is famous and known for.

Seng Kee Black chicken herbal soup- Changi Rd, Singapore — To try their pig kidney mee sua, steamed fish, sambal petai, lemon clams, claypot sesame chicken.

While this is their main outlet at Kembangan , they also have a branch in Bedok hawker centre so you can try both the sambal stingray and the signature pig kidney mee sua from the same market. Whampoa Keng Fish Head Steamboat Restaurant Rangoon Rd, Singapore — To try their fish head steamboat addictive, tasty soup heated over charcoal- could ask for refill too , homemade tofu.

Eng Seng restaurant- Joo Chiat Place, for their black pepper crab. Do call to reserve for the crabs before heading there. Known to serve the best black pepper crab in Singapore amongst the locals. Melben Seafood To try their crabs, prawn paste chicken, Homemade Beancurd. Welcome in advanced to Singapore! My family and I have been going to Bangkok quite frequently so we sometimes turn to you for suggestions on places to eat in Bangkok which are always super amazing , so I really hope this helps you for your trip to Singapore!

Try the maoshanwang durian! Though durian is not in season now.. For Chilli Crab — Jumbo Seafood. You have to wait up to 40 mins for the Hokkien Mee so just a heads up for you. The queue during peak hours gets quite crazy though so be sure to avoid the crowd. The sauces for their dishes are absolutely amazing. The Lor Mee stall is called Lor Mee at We have been looking forward to this day!

Do let us know if a meetup is organised! Laksa — Sungei road laksa; Hands down the best laksa imho! Bak Kut Teh — As you may be aware, there are two types of bak kut teh; pepper and herbal base Pepper: Ah Hock Hokkien Mee http: Boon Tong Kee Balestier http: Wan Ton Mee — Poor staff attitude; long waiting time; absolutely delicious wanton mee! Fish Bee Hoon Not a fan! Bak Chor Mee — Hands down the best!!! Gotta be there early before some types of the noodles run out, esp mee kia and mee pok http: Oyster Omelet Orh Lua — hidden gem!

Blk Ang Mo Kio Ave 10 Xi Xiang Feng Youn http: Roast Duck — It also has my all time favourite char siew!! Peranakan Food Not a big fan! They also have my favourite curry fish head! I sincerely hope you try this! Rojak — their fried dough fritters are toasted; perfect shrimp paste sauce with heaps of ground peanuts! Ice Kachang Not a big fan! Nasi Lemak — Online review may not be great due to exorbitant pricing as well as staff attitude, however, definitely my all time favourite!

Die die must try!! Fish Head Curry See Singapore known for multiracial country. HI Mark, first and foremost, welcome to Singapore — the food haven for all foodies! Going to recommend some of my personal favourites.

Bak Kut Teh — generally there are 2 types. The teochew version which is clear broth cooked with garlic and white peppercorns — so if you like spicy, this would be the one and the hokkien version which is a dark broth cooked with all sorts of chinese herbs, so it taste more on the sweet side. My personal favourite is the teochew version, and I like it at this stall: On weekends, do go early — they open about 5pm.

If you come at about 5. It is in a hawker centre aka open air and sheltered food court. Carrot Cake Chai Tow Kway — again, there are 2 ways of frying black and white. Black is one where it is pan-fried with sweet sauce so it tastes both sweet and salty.

White is one where is just pan-fried with eggs. Check out this link to see the food-stall picture: Likely they will start to close in the afternoon. The noodles are springy and nice. At busy lunch hours including weekends, where people eat brunch , the queue may be 45 mins or longer. Ask anyone in Singapore about Kay Lee which reportedly sold its secret recipe for 4 million Singapore Dollars, and most will have recollection of it. The meats here are tender, and roast till crispy and perfection.

The duck is fabulously good. Do make an advance reservation because it may get packed even on weekdays. When calling them to reserve a place for you, you would have to also inform the person who picks up your call, that you also want to reserve a Peking Duck — otherwise you may be disappointed.

Peranakan Food — I am a Peranakan which means that my ancestors are from inter-racial marriages, ie. You would need to take a taxi into the club though, because it is very far in from the main road. You would definitely have to try the Ayam Buah Keluak, which is a signature Peranakan dish.

You may surf their website for more. The address is http: My personal favourite is Kimly Live Seafood 18 Boon Lay Way, Tradehub 21, , Singapore — they are located like in a open air and sheltered food court. They used Sri Lanka crabs which is more firm and fleshy. View this image to get an idea of the crab mentioned above: The snaking queues is present no matter what time of the day be prepared for at least 30 mins wait.

This means that the prawn paste used is of a high quality. Durian — the most area which sells this is in Geylang. But sometimes they sell at super exorbitant prices — esp. For durian lovers, they would say you have to go for Malaysia durians, rather than the Thailand ones sorry! Hope the above helps you in your food choice! Keep in touch if you need more help.. Chomp Chomp has quite a handful of stalls, can just order from stalls which are more bustling. Contact me, I will be your host for a day and drive you across the causeway to Johor Bahru to try some delicacies.

Hi Mark, hope the following would be useful for you! Hopefully these recommendations will suit your taste buds!

Would love to see a video should you try any one of these! Blk Ubi Road 1, , Singapore Operating hours: That restaurant was the pioneer of chili crab in Singapore, trust me! Laksa — Katong Laksa 2. Chili Crab — Jumbo Fish Curry — Muthu Curry They only sell one type other modern stalls would sell soup and dry noodles and allow you to pick your own toppings.

The stallholder will typically ask customers if they want bittergourd to be included. In my opinion, just say yes and try it. Nearest MRT station would be Bugis. Anyway, try to avoid touristy areas usually recommended on tour websites!! Then go to East coast lagoon food village!!! Hi Mark, you must try: FIRST of all, please take this comment of mine seriously and really heed the following advice I have to give!

Second of all, you have to use one and only one local Sg blog. He literally set up the blog as a practical means for other clueless Singaporeans to use it as a catalogue of sorts to trawl for local eats, back in the older days of the blog.

It has now flourished into a serious hobby of his. Dr Tay would know. A bit of sg good background. Do a quick google. Then, Char Kueh Teow must only be eaten at the stall from Hillstreet. And of course white skinned not roasted Chicken Rice only from Maxwell — the one that Gordon Ramsay himself got involved in a little competition with not too long ago. The same applies to every other dish you have listed. Go to the best. He blogs about it annually during durian season and advocates durian seller honesty hahaha.

Perhaps compare the durians sold here with the ones back home? You can delve deeper into his blog for specific things if you want to. I understand you initiated this feedback thing where we suggest to you, because you wanna make the most out of your trip in Singapore yeah? And well I wanna proudly give you and Ying the best impression of my country.

Hence my totally honest answer: For comprehensive list of the best stalls to visit. We try to improve both localization and recognition. On the localization side, initial candidate proposals are generated from selective search [1], and a novel bounding boxes regression method is used for better object localization.

Given these features, category-specific combination functions are learnt to improve object recognition. Background priors and object interaction priors are also learnt and applied into our system. In addition, our framework involves some other novel techniques.

The pertinent technical details for the submission are in preparation. Selective search for object recognition[J]. International journal of computer vision, , 2: Rich feature hierarchies for accurate object detection and semantic segmentation[J].

Springer Berlin Heidelberg, We train several 6-layer convnets using ImageNet classes for classification and then adapt one model for bounding box regression. At test time, we use k-means to find bounding box clusters and rank the clusters according to the classification scores. Andrew Howard Andrew Howard - Howard Vision Technologies Deep convolutional neural networks are very costly to train so my submission focuses on reusing networks through retraining and by using the same network to make multiple predictions.

The differences from the base ZF model are that I use 7 convolutional layers with convolutional layer having filters. It took over 6 weeks to train on a GTX Titan using cuda-convnet [2].

This base model is trained using x crops from the full xN image [3] with random horizontal flips [4]. Each training crop is further perturbed with color channel noise [4] and random variation in photometric properties lighting,contrast,color [3].

This base model is then adapted to build a high resolution [3] and a low resolution model. The low resolution model embeds the entire image resized to xN into a random location in the x crop for retraining.

I also retrain the base model to increase the size of the fully connected layers to a size larger than would fit in GPU memory if the model was trained together the fully connected layer is grown from x to x and trained from scratch while the convolutional layers are held fixed.

When the new fully connected layers are retrained, I use a slow form of Polyak averaging which averages the model parameters after each epoch rather than after each iterate.

Each of the 3 models are responsible for 2 resolutions. The base resolution model acts on images scaled at xN and xN. The high resolution model acts on xN and xN and the low resolution acts on xN and xN.

Each resolution uses locations selected on a dense spatial grid on the entire image similar to [5]. Predictions at each spatial location are averaged into a prediction for a given resolution and then predictions are each resolution are combined evenly.

For features, I use the final dimension aggregate predictions. I use leave one out cross validation on the validation set to choose K the number of neighbors and the weighting between the final neural network prediction and the KNN prediction. Finally I adapt the neural networks to the validation set distribution as suggested by the NUS team last year [6]. To do this, I hold fixed the convolutional layers and adapt the fully connected layers to the validation set. The final submission is made up of 2 sets of 3 networks plus 1 KNN prediction.

The second set of networks are a smaller earlier version and only add a little value. Fergus, "Visualizing and Understanding Convolutional Networks.

Hinton, "Imagenet classification with deep convolutional neural networks. Integrated recognition, localization and detection using convolutional networks. Each resolution gets even weighting in the final prediction. Institute for Infocomm Research, Singapore 2: The deep representations are extracted across multiple scales and positions within an image. Model fusion is adaptively performed within each CNN model, and subsequently across the different models.

Class distribution priors are used to rectify the outputs of the model. We submit three runs for the classification-only task. No external data was used in our models. A single CNN model. Adaptive fusion of multiple CNN models. Adaptive fusion of multiple CNN models with output rectification. In this network, convolutional layers one through five have 96, , , , and filters, respectively. The two fully connected layers before the linear classifiers both have output units.

We performed control experiments to compare these changes to the results in [2]. On the val2 validation set see [2] , the new training data added for improved results from Using the slightly larger CNN improved results on val2 to Bounding-box regression further increased this to Rich feature hierarchies for accurate object detection and semantic segmentation.

The overall training details are based on [2]. We used caffe[3] as our development environment. For localization, we computed image specific class saliency as in [4]. Delving Deep into Convolutional Nets. An open source convolutional architecture for fast feature embedding.

However, the object proposals are different from those used in R-CNN, explained as follows. Another difference is that, to obtain the pre-training CNN model, we use the categories images on dataset 1 for training rather than the categories images on dataset 2. Then, a multiple instance learning strategy is adopted to learn the object detectors with weak supervision.

Using the learned object detectors, we are able to learn object classifiers instead of global image classifiers using multi-class softmax model.

Finally, the detection models and classification models are fused to produce the final classification results. Simple reweighting techniques are used as well.

Our localization method is a bounding box regression. In order to train a global expert, we have used 10 networks under different settings: We selected the best ensemble of the networks that demonstrate the best accuracy in the validation dataset.

Our local expert is trained using local features composed of DCNN responses from mid-layers. We encoded the local features into Fisher vectors [2] and trained SVM classifiers. In order to prevent overfitting, we trained our network using 0. This detection work is based on deep CNN with proposed new deformation layers, feature pre-training strategy, sub-region pooling and model combination. The effectiveness of learning deformation models of object parts has been proved in object detection by many existing non-deep-learning detectors, e.

However, it is missed in current deep learning models. In deep CNN models, max pooling and average pooling are useful in handling deformation but cannot learn the deformation penalty and geometric model of object parts. We design the deformation layer for deep models so that the deformation penalty of objects can be learned by deep models.

The deformation layer was first proposed in our recently published work [b], which showed significant improvement in pedestrian detection. In this submission, we extend it to general object detection on ImageNet. In [b], the deformation layer was only applied to a single level corresponding to body parts, while in this work the deformation layer was applied to every convolutional layer to capture geometric deformation at all the levels.

In [b], it was assumed that a pedestrian only has one instance of a body part, so each part filter only has one optimal response in a detection window. In this work, it is assumed that an object has multiple instances of body part e. This new model is more suitable for general object detection. The whole detection pipeline is much more complex than [b]. In addition to the above improvement, we also added several new components in the pipeline, including feature pre-training on the ImageNet classification dataset objective function is the image classification task , feature fine-tuning on the ImageNet detection dataset objective function is the object detection task , a proposed new sub-region pooling step, contextual modeling which uses the whole image prediction scores over classes as contextual features to combine with features extracted from a detection window with deep CNN , SVM classification by using the extracted features.

We also adopted bounding box regression [c]. A new sub-region pooling strategy is proposed. It divides the detection window into sub-regions, and applies max-pooling or average pooling across feature vectors extracted from different sub-regions. It improves the performance and also increases the model diversity. Different from the state-of-the-art deep learning detection framework [c], which pretrain the net on ImageNet classification data classes , We proposed a new strategy of doing pre-training on the ImageNet classification data classes , such that the pre-trained features are much more effective on the detection task and with better discriminative power on object localization.

By changing the configuration of each component of the detection pipeline, multiple models with large diversity are generated. Multiple models are automatically selected and combined to generate the final detection result. We have submitted the results of five different approaches. The first two results report the best performance to be achieved with a single model. Their difference is whether using contextual features from image classification or not.

The remaining three results report the best performance to be achieved with model combination. Their differences are using contextual modeling or not, and whether using validation 2 dataset from ImageNet as part of training or not.

Object detection with discriminatively trained part based models. Malik, "Rich feature hierarchies for accurate object detection and semantic segmentation", CVPR This detection work is based on multi-stage deep CNN and model combination. Multi-Stage classifiers have been widely used in object detection and achieved great success. With a cascaded structure, each classifier processes a different subset of data.

However, these classifiers are usually trained sequentially without joint optimization. In this submission, we proposed a new deep architecture that can jointly train multiple classifiers through several stages of back-propagation. Each stage handles samples at a different difficulty levels. Through a specific design of the training strategy, this deep architecture is able to simulate the cascaded classifiers by mining hard samples to train the network stage-by-stage.

The group of classifiers in the deep model choose training samples stage by stage. The training is split into several back-propagation BP stages. Due to the design of our training procedure, the gradients of classifier parameters at the current stage are mainly influenced by the samples misclassified by the classifiers at the previous stages. At each BP stage, the whole deep model has been initialized with a good starting point learned at the previous stage and the additional classifiers focus on the misclassified hard samples.

Direct back-propagation on the multi-stage deep CNN easily lead to the overfitting problem. We design stage-wise supervised training to regularize the optimization problem. At each BP stage, classifiers at the previous stages jointly work with the classifier at the current stage in dealing with misclassified samples.

Existing cascaded classifiers only pass a single score to the next stage, while our deep model keeps the score map within a local region and it serves as contextual information to support the decision at the next stage. Our recent work [1] has explored the idea of multi-stage deep learning, but it was only applied to pedestrian detection. In this submission, we apply it to general object detection on ImageNet. The detection pipeline is much more complex than [1].

It includes feature pre-training, multi-stage deep CNN fine-tuning, sub-region pooling, contextual modeling , SVM classification, and bounding box regression. The state-of-the-art deep learning object detection framework in [2] pretrain the net on ImageNet classification data classes and then fine-tune on ImageNet detection data classes. We proposed a new strategy of doing pre-training on the ImageNet classification data classes , such that the pre-trained features are much more effective on the detection task and with better discriminative power on object localization.

Context modeling uses the whole image prediction scores over classes as contextual features to combine with features extracted from a detection window with deep CNN. By changing the configuration of each step, we can generate multiple deep models. With extracted features, bounding boxes can be classified with fully connected networks with hinge loss or SVM, including sub-region pooling or not.

Therefore, different models can be generated. Top N models with the highest accuracies are combined by averaging. The work uses ImageNet classification training set classes to pre-train features, and fine tunes features on ImageNet detection training set classes. No other training data is used. In our entry, we use the selective search and structure edge to generate around object proposals for each image.

The features of each object proposal are extracted from three CNNs, which are trained on the classification task and tuned on the detection task. The three CNNs are different in the depth of convolution layer.

Deeper model always achieves better result according to the validation set. The bounding box regression uses the output of the final layer as the input to refine the result. For the context, we train binary classifiers on the detection data and use them to re-score the detection.

Multi-column deep neural networks for image classification, Sparsity is used to accelerate the training process Graham, Sparse arrays of signatures for online character recognition http: For training, affine transformations are used. For testing, each image is fed forward through the network only once.

Regarding Q3 in the FAQ "Do teams have to submit both classification and localization results in order to participate in Task 2? Thank you to all the organisers. To fully optimize such a deep model, we adopt a Nesterov based optimization method which is shown to be superior to the conventional SGD. We also exploit more advanced data augmentation technique such as using various resolution, lightness and contrast variation, etc.

For model ensemble, we directly use discrete optimization to optimize top 5 error rate. For non-square images, instead of using central crop which in many cases, does not contain the object of interest at all or the object is incomplete , we apply Graph-Based Visual Saliency by Harel et al. NIPS to the original image both in training and testing and use integral image to get a square crop that maximizes the visual saliency.

One of the two submissions is from a single CNN. The other combines multiple CNNs. Additional dimension reduction layers based on embedding learning intuition allow us to increase both the depth and the width of the network significantly without incurring significant computational overhead. Combining these ideas allow for increasing the number of parameters in convolutional layers significantly while cutting the total number of parameters and resulting in improved generalization.

Various incarnations of this architecture are trained for and applied at various scales and the resulting scores are averaged for each image. In particular, we want to compare the results of different algorithms that can produce region proposals, and to find out which is the most important factor that influence the following classification.

Zeiler, and Rob Fergus trained with 1 convnet, detailed in: For the fisher based model [1], fisher vectors were extracted using local feature descriptors.

Linear classifiers were trained for these fisher vectors using the averaged passive-aggressive algorithm. The classifier was trained using [2] which trains a multiclass classifier by creating 'negative classes' for each class. This optimises the separation between positive and negative features while simultaneously optimising the separation between classes.

We use RCNN[3] as the base detector. We train separate fisher based classifiers for each class using the Passive Aggressive algorithm. The scores from these classifiers for each image is collected and is used for rescoring the detections. For implementing TR, we extracted dimensional global CNN features by Caffe [5] and retrieved nearest training samples in terms of L2 distance.

We computed to windows for each object region proposal method and then put them together for RCNN. Using pre-trained CNN and SVM models provided by RCNN software, we computed scores for each proposal and ran non-maxima suppression without distinguishing proposal methods to determine the final predictions.

We did not perform bounding box regression refinement as the original RCNN paper does. We observed that combining different object proposal methods worked better than just computing more proposals by one method. The SPP-net is also a fast and accurate solution to object detection. We compute the convolutional feature maps from the images only once, and use SPP to pool features from arbitrary proposal windows for training SVM detectors. Our method is tens of times faster than R-CNN.

A few strategies are proposed to improve the pre-training, driven by the different statistical properties of the DET set. The algorithm details have been described in our ECCV paper.

An extended technical report will be updated. The code will be released.