Learning methods for generic object recognition with invariance to pose and lighting

By Yann LeCun, Fu Jie Huang, L. Eon Bottou. 2004

1,593 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

OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks (2013, background): “Recognizing the category of the dominant object in an image i s a tasks to which Convolutional Networks (ConvNets) [17] have been applied for many years, w hether the objects were handwritten characters [16], house numbers [24], textureless toys [18] , traffic signs [3, 26], objects from the Caltech”
Pylearn2: a machine learning research library (2013, methodology): “These include the datasets used for DARPA’s unsupervised and transfer learning challenge [25] , the dataset used for the NIPS 2011 workshops challenge in representation learning [32], the C IFAR-10 and CIFAR-100 datasets [29], the MNIST dataset [34], some of the MNIST variations dataset s [31], the ”
On Robustness and Transferability of Convolutional Neural Networks (2020, background): “The structured group contains the following tasks: counting and distance prediction on CLEVR [33], pixel-location and orientation prediction on dSprites [45], camera elevation and object orientation on SmallNORB [41], object distance on DMLab [4] and vehicle distance on KITTI [19].”
A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmark (2019, background): “SmallNORB/azimuth (LeCun et al., 2004) The Small NORB dataset contains images of 3D-toys from 50 classes, including animals, human ﬁgures, airplanes, trucks, and cars.”
A Sober Look at the Unsupervised Learning of Disentangled Representations and their Evaluation (2020, methodology): “On all data sets except SmallNORB, we observe that plain vanilla variational autoencoders (the β-VAE model with β = 1) exhibit the highest total correlation.”
Learning from Simulated and Unsupervised Images through Adversarial Training (2016, background): “various prediction tasks, including gaze estimation [43], text detection and classification in RGB images [9, 15], font recognition [42], object detection [10, 27], hand pose estimation in depth images [38, 37], scene recognition in RGB-D [11], semantic segmentation of urban scenes [31], and human p”
Deep Speech 2 : End-to-End Speech Recognition in English and Mandarin (2015, background): “Data augmentation has been highly effective in improving the performance of deep learning in computer vision (LeCun et al., 2004; Sapp et al., 2008; Coates et al., 2011) and speech recognition (Gales CTC”
Visual Prompt Tuning (2022, neutral): “800/1000 200 15,000 Clevr/distance [34] 6 15,000 DMLab [3] 6 22,735 KITTI/distance [20] 4 711 dSprites/location [53] 16 73,728 dSprites/orientation [53] 16 73,728 SmallNORB/azimuth [40] 18 12,150 SmallNORB/elevation [40] 9 12,150”
Multi-column deep neural networks for image classification (2012, methodology): “Here we train such nets by simple online back-propagation, setting new, greatly improved records on MNIST [19], Latin letters [13], Chinese characters [22], traffic signs [33], NORB (jittered, cluttered) [20] and CIFAR10 [17] benchmarks.”
Auxiliary Deep Generative Models (2016, background): “We normalized all NORB images following Miyato et al. (2015) using image pairs of 32 x 32 resulting in a vectorized input of dim( x ) = 2048 .”
Deep Speech: Scaling up end-to-end speech recognition (2014, methodology): “To expand our potential training data even further we use data synthesis, which has been successfully applied in other contexts to amplify the effective number of training samples [37, 26, 6].”
Learning Image Representations Tied to Ego-Motion (2015, methodology): “The KITTI data itself is vastly more challenging than NORB due to its noisy ego-poses from inertial sensors, dynamic scenes with moving traffic, depth variations, occlusions, and objects that enter and exit the scene.”
Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learning (2016, background): “It is shown that this loss function makes use of unlabeled data and pushes the decision boundary to a less dense area of decision space.”
LabelMe: A Database and Web-Based Tool for Image Annotation (2008, background): “There are a large number of publically available databases of visual objects (Torralba et al. 2004; Agarwal et al. 2004; Leibe et al. 2004; Opelt et al. 2006b; Everingham et al. 2006; Fei-Fei et al. 2004, 2007; Griffin et al. 2006; Carmichael and Hebert 2004; Li and Shapiro 2002; LeCun et al. 2004; ”
Ladder Variational Autoencoders (2016, methodology): “To test our models we use the standard benchmark datasets MNIST, OMNIGLOT [10] and NORB [11].”
On the Transfer of Disentangled Representations in Realistic Settings (2020, methodology): “The SmallNORB dataset contains toy images rendered under diﬀerent lighting conditions, elevations and azimuths (LeCun et al., 2004).”
Weakly-Supervised Disentanglement Without Compromises (2020, background): “…noise; we assume that noise is explicitly modeled as a latent variable and its effect is manifested through g? as done by (Bengio et al., 2013; Locatello et al., 2019b; Higgins et al., 2018; 2017a; Suter et al., 2019; Reed et al., 2015; LeCun et al., 2004; Kim & Mnih, 2018; Gondal et al., 2019).”
Weakly Supervised Disentanglement with Guarantees (2019, methodology): “We conducted experiments on ﬁve prominent datasets in the disentanglement literature: Shapes3D (Kim & Mnih, 2018), dSprites (Higgins et al., 2017), Scream-dSprites (Locatello et al., 2019), Small-NORB (LeCun et al., 2004), and Cars3D (Reed et al., 2015).”
On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Dataset (2019, methodology): “every 5 degrees), and 18 azimuths (0 to 340 every 20 degrees) [27] and the data is available at https://cs.”