Deep Residual Learning for Image Recognition
Kaiming He,Xiangyu Zhang,Shaoqing Ren,Jian Sun +3 more
- pp 770-778
TLDR
In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.Abstract:
Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.read more
Citations
More filters
Posted Content
Taming Transformers for High-Resolution Image Synthesis
TL;DR: It is demonstrated how combining the effectiveness of the inductive bias of CNNs with the expressivity of transformers enables them to model and thereby synthesize high-resolution images.
Book ChapterDOI
The Visual Object Tracking VOT2016 Challenge Results
Matej Kristan,Ales Leonardis,Jiří Matas,Michael Felsberg,Roman Pflugfelder,Luka Cehovin,Tomas Vojir,Gustav Häger,Alan Lukežič,Gustavo Fernandez,Abhinav Gupta,Alfredo Petrosino,Alireza Memarmoghadam,Alvaro Garcia-Martin,Andres Solis Montero,Andrea Vedaldi,Andreas Robinson,Andy J. Ma,Anton Varfolomieiev,A. Aydin Alatan,Aykut Erdem,Bernard Ghanem,Bin Liu,Bohyung Han,Brais Martinez,Chang-Ming Chang,Changsheng Xu,Chong Sun,Daijin Kim,Dapeng Chen,Dawei Du,Deepak Mishra,Dit-Yan Yeung,Erhan Gundogdu,Erkut Erdem,Fahad Shahbaz Khan,Fatih Porikli,Fatih Porikli,Fei Zhao,Filiz Bunyak,Francesco Battistone,Gao Zhu,Giorgio Roffo,Gorthi R. K. Sai Subrahmanyam,Guilherme Sousa Bastos,Guna Seetharaman,Henry Medeiros,Hongdong Li,Honggang Qi,Horst Bischof,Horst Possegger,Huchuan Lu,Hyemin Lee,Hyeonseob Nam,Hyung Jin Chang,Isabela Drummond,Jack Valmadre,Jae-chan Jeong,Jaeil Cho,Jae-Yeong Lee,Jianke Zhu,Jiayi Feng,Jin Gao,Jin-Young Choi,Jingjing Xiao,Ji-Wan Kim,Jiyeoup Jeong,João F. Henriques,Jochen Lang,Jongwon Choi,José M. Martínez,Junliang Xing,Junyu Gao,Kannappan Palaniappan,Karel Lebeda,Ke Gao,Krystian Mikolajczyk,Lei Qin,Lijun Wang,Longyin Wen,Luca Bertinetto,Madan Kumar Rapuru,Mahdieh Poostchi,Mario Edoardo Maresca,Martin Danelljan,Matthias Mueller,Mengdan Zhang,Michael Arens,Michel Valstar,Ming Tang,Mooyeol Baek,Muhammad Haris Khan,Naiyan Wang,Nana Fan,Noor M. Al-Shakarji,Ondrej Miksik,Osman Akin,Payman Moallem,Pedro Senna,Philip H. S. Torr,Pong C. Yuen,Qingming Huang,Qingming Huang,Rafael Martin-Nieto,Rengarajan Pelapur,Richard Bowden,Robert Laganiere,Rustam Stolkin,Ryan Walsh,Sebastian B. Krah,Shengkun Li,Shengping Zhang,Shizeng Yao,Simon Hadfield,Simone Melzi,Siwei Lyu,Siyi Li,Stefan Becker,Stuart Golodetz,Sumithra Kakanuru,Sunglok Choi,Tao Hu,Thomas Mauthner,Tianzhu Zhang,Tony P. Pridmore,Vincenzo Santopietro,Weiming Hu,Wenbo Li,Wolfgang Hübner,Xiangyuan Lan,Xiaomeng Wang,Xin Li,Yang Li,Yiannis Demiris,Yifan Wang,Yuankai Qi,Zejian Yuan,Zexiong Cai,Zhan Xu,Zhenyu He,Zhizhen Chi +140 more
TL;DR: The Visual Object Tracking challenge VOT2016 goes beyond its predecessors by introducing a new semi-automatic ground truth bounding box annotation methodology and extending the evaluation system with the no-reset experiment.
Posted Content
CornerNet: Detecting Objects as Paired Keypoints
TL;DR: CornerNet, a new approach to object detection where an object bounding box is detected as a pair of keypoints, the top-left corner and the bottom-right corner, using a single convolution neural network, is proposed.
Posted Content
The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization
Dan Hendrycks,Steven Basart,Norman Mu,Saurav Kadavath,Frank Wang,Evan Dorundo,Rahul Desai,Tyler Zhu,Samyak Parajuli,Mike Guo,Dawn Song,Jacob Steinhardt,Justin Gilmer +12 more
TL;DR: It is found that using larger models and artificial data augmentations can improve robustness on real-world distribution shifts, contrary to claims in prior work.
Journal ArticleDOI
Wide Neural Networks of Any Depth Evolve as Linear Models Under Gradient Descent
Jaehoon Lee,Lechao Xiao,Samuel S. Schoenholz,Yasaman Bahri,Roman Novak,Jascha Sohl-Dickstein,Jeffrey Pennington +6 more
TL;DR: In this article, the authors show that for wide neural networks the learning dynamics simplify considerably and that, in the infinite width limit, they are governed by a linear model obtained from the first-order Taylor expansion of the network around its initial parameters.
References
More filters
Proceedings Article
ImageNet Classification with Deep Convolutional Neural Networks
TL;DR: The state-of-the-art performance of CNNs was achieved by Deep Convolutional Neural Networks (DCNNs) as discussed by the authors, which consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connected layers with a final 1000-way softmax.
Journal ArticleDOI
Long short-term memory
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Proceedings ArticleDOI
Going deeper with convolutions
Christian Szegedy,Wei Liu,Yangqing Jia,Pierre Sermanet,Scott Reed,Dragomir Anguelov,Dumitru Erhan,Vincent Vanhoucke,Andrew Rabinovich +8 more
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
Proceedings Article
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Sergey Ioffe,Christian Szegedy +1 more
TL;DR: Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.