ImageNet Large Scale Visual Recognition Challenge
By Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael S. Bernstein and 2 more. 2014
43,116 citations recorded by the source corpus.
The ImageNet Large Scale Visual Recognition Challenge is a benchmark in object category classification and detection on hundreds of object categories and millions of images. The challenge has been run annually from 2010 to present, attracting participation from more than fifty institutions. This paper describes the creation of this benchmark dataset and the advances in object recognition that have been possible as a result. We discuss the challenges of collecting large-scale ground truth annotation, highlight key breakthroughs in categorical object recognition, provide a detailed analysis of the current state of the field of large-scale image classification and object detection, and compare the state-of-the-art computer vision accuracy with human accuracy. We conclude with lessons learned in the 5 years of the challenge, and propose future directions and improvements.
Subfields: Computer Vision and Pattern Recognition, Evaluation & Benchmarks
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Luc Van Gool
- Dacheng Tao
- Ming-Hsuan Yang
- Trevor Darrell
- Xiaogang Wang
- Jitendra Malik
- Andrew Zisserman
- Yi Yang
In Evaluation & Benchmarks
- Mohit Bansal
- Luc Van Gool
- Trevor Darrell
- Jitendra Malik
- Li Fei-Fei
- Vladlen Koltun
- Antonio Torralba
- Abhinav Kumar Gupta
Citations in the corpus
- Xception: Deep Learning with Depthwise Separable Convolutions (2016, methodology): “Since its first introduction, Inception has been one of the best performing family of models on the ImageNet dataset [14], as well as internal datasets in use at Google, in particular JFT [5].”
- Rethinking the Inception Architecture for Computer Vision (2015, methodology): “th using less than 25 million parameters. With an ensemble of 4 models and multi-crop evaluation, we report 3:5% top-5 error and 17:3% top-1 error. 1. Introduction Since the 2012 ImageNet competition [16] winning entry by Krizhevsky et al [9], their network “AlexNet” has been successfully applied to”
- Learning Distributed Representations of Sentences from Unlabelled Data (2016, methodology): “, 2014) trained on the ILSVRC 2014 object recognition task (Russakovsky et al., 2014).”
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (2015, methodology): “, the shared convolutional layers) are initialized by pre-training a model for ImageNet classification [36], as is standard practice [5].”
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (2015, methodology): “gresses, which aids the training. 4.2 ImageNet classification We applied Batch Normalization to a new variant of the Inception network Szegedy et al. (2014), trained on the ImageNet classification task Russakovsky et al. (2014). The network has a large number of convolutional and pooling layers, with ”
- FitNets: Hints for Thin Deep Nets (2014, methodology): “Similarly, we extracted 25K 16x16 pixels patches not containing faces from ImageNet (Russakovsky et al., 2014) dataset, as negative samples.”
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition (2014, background): “These spatial bins have sizes proportional to the image size, so the number of bins is fixed regardless of the image size.”
- Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer (2019, background): “In applications of transfer learning to computer vision [Oquab et al., 2014; Jia et al., 2014; Huh et al., 2016; Yosinski et al., 2014], pre-training is typically done via supervised learning on a large labeled dataset like ImageNet [Russakovsky et al., 2015; Deng et al., 2009].”
- A Simple Framework for Contrastive Learning of Visual Representations (2020, positive): “Most of our study for unsupervised pretraining (learning encoder network f without labels) is done using the ImageNet ILSVRC-2012 dataset (Russakovsky et al., 2015).”
- Adversarial NLI: A New Benchmark for Natural Language Understanding (2019, background): “Progress in AI has been driven by, among other things, the development of challenging large-scale benchmarks like ImageNet (Russakovsky et al., 2015) in computer vision, and SNLI (Bowman et al.”
- A Constructive Prediction of the Generalization Error Across Scales (2019, positive): “…all scales investigated and under 5% when extrapolating from a slimmed-down model (1/16 of the parameters) on a fraction of the training data (1/8 of the examples) on the ImageNet (Russakovsky et al., 2015) and WikiText-103 (Merity et al., 2016) datasets, with similar results for other datasets.”
- Deep Learning Scaling is Predictable, Empirically (2017, methodology): “We test ResNets (He et al. (2016)), which were recently the SOTA architectures for ImageNet clas-sification (Russakovsky et al. (2015)).”
- Rethinking ImageNet Pre-Training (2018, background): “, ImageNet [39]) and then to fine-tune the models on target tasks that often have less training data.”