Large Scale Visual Recognition

By Jia Deng. 2012

89 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift (2015, methodology): “This improves upon the previous best result, and exceeds the estimated accuracy of human raters according to (Russakovsky et al., 2014).”
The Many Faces of Robustness: A Critical Analysis of Out-of-Distribution Generalization (2020, neutral): “” (Deng, 2012).”
Do ImageNet Classifiers Generalize to ImageNet? (2019, background): “To be precise: Jia Deng’s PhD thesis [9] provides a clear high-level description of their algorithm for computing agreement thresholds.”
Movie Description (2016, methodology): “We use both available networks, Places-CNN and Hybrid-CNN , where the ﬁrst is trained on the Places dataset (Zhou et al. 2014) only, while the second is additionally trained on the 1.2 million images of ImageNet (ILSVRC 2012) (Russakovsky et al. 2014).”
Towards Open World Recognition (2014, methodology): “Finally, we present a protocol for evaluation for open world recognition, and use this protocol to show our NNO algorithm perform significantly better on open world recognition evaluation using Image-Net [1].”
Towards Practical Verification of Machine Learning: The Case of Computer Vision Systems (2017, methodology): “…the state-of-the-art performances in ILSVRC [42] competitions; (2) ﬁve commercial third-party vision APIs provided by Google [18], Clarifai [13], IBM [25], Microsoft [30], and Amazon [4]; (3) four popular self-driving car models including Rambo, one of the top performers in Udacity…”
K-Means Hashing: An Affinity-Preserving Quantization Method for Learning Binary Compact Codes (2013, background): “A discussion on the two terminologies of “hashing” can be found in [24].”
Learning Features and Parts for Fine-Grained Recognition (2014, methodology): “This architecture was determined by extensive experiments using a validation set drawn from the training set, as the network used in [25] suffers from substantial overfitting on fine-grained datasets, which typically have over 100x less training data than in ILSVRC2012 [39], the dataset used to trai”
Detecting Avocados to Zucchinis: What Have We Done, and Where Are We Going? (2013, methodology): “An alternative is to use the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) [3] data with 1,000 object classes for benchmarking and analyzing detection.”
From Large Scale Image Categorization to Entry-Level Categories (2013, background): “Although far from solved, algorithms have now advanced to the point where they can recognize or localize thousands of object categories with reasonable accuracy [17, 4, 3, 12].”
Meta-class features for large-scale object categorization on a budget (2012, background): “Thanks to the compactness of this representation, it is possible to keep the entire ILSVRC2010 database in the memory of a budget PC, and train classifiers for all 1000 classes in 14 hours on a single core machine.”
Stochastic Multiple Choice Learning for Training Diverse Deep Ensembles (2016, background): “For example, in classiﬁcation tasks, oracle accuracy is exactly the top-k criteria of ImageNet[25], i.e . whether at least one of the outputs is the correct label.”