Synergistic Face Detection and Pose Estimation with Energy-Based Models

By Margarita Osadchy, Yann LeCun, Matthew L. Miller. 2004

431 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

Strike (With) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects (2018, methodology): “), we parameterized each rotation angle θi as (cos(θi), sin(θi))—a technique commonly used for pose estimation [30] and inverse kinematics [11]—which maps the Cartesian plane to angles via the atan2 function.”
DeepPose: Human Pose Estimation via Deep Neural Networks (2013, methodology): “[17] employ a NN-based pose embedding trained with a contrastive loss.”
Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture (2014, neutral): “However, ConvNets have recently been applied to a variety of other tasks, including pose estimation [36, 27], stereo depth [38, 25], and instance segmentation [14].”
Deep learning for detecting robotic grasps (2013, background): “During inference , our goal is to ﬁnd the single grasping rectangle with the maximum probability of being graspable for some new object.”
Flowing ConvNets for Human Pose Estimation in Videos (2015, background): “The power of ConvNets has been demonstrated in a wide variety of vision tasks – object classification and detection [11, 21, 28, 40], face recognition [32], text recognition [15, 16], video action recognition [20, 30] and many more [7, 22, 25].”
Learning Human Pose Estimation Features with Convolutional Networks (2013, neutral): “applied a convolutional network to simultaneously detect and estimate the pitch, yaw and roll of a face [31].”
Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition (2010, background): “CNNs have achieved state-of-the-art results for the recognition of handwritten digits [23] and for the detection of faces [17].”
Head pose estimation in the wild using Convolutional Neural Networks and adaptive gradient methods (2017, positive): “Our approach is significantly different from previous research [18, 20, 21, 17], and show how using the most recent 605 deep learning techniques leads to the state-of-the-art in constrained and unconstrained datasets.”
Real time head pose estimation with random regression forests (2011, background): “[21], for example, use a convolutional network, detecting faces and their orientation in real-time.”
Large-Scale FPGA-based Convolutional Networks (2011, background): “, 2006), • detection in images, including faces with record accuracy and real-time performance (Vaillant et al., 1994; Garcia and Delakis, 2004; Osadchy et al., 2007; Nasse et al., 2009), license plates and faces in Google’s StreetView (Frome et al.”
Non-Stationary Markov Decision Processes a Worst-Case Approach using Model-Based Reinforcement Learning (2019, background): “In this matter, the necessity to have access to a model is a great concern of AI (Asadi et al., 2018; Osadchy et al., 2007; Jaulmes et al., 2005; Doya et al., 2002; Da Silva et al., 2006).”
Face Detection with the Faster R-CNN (2016, methodology): “In [22], a CNN is trained to perform joint face detection and pose estimation.”
From Facial Parts Responses to Face Detection: A Deep Learning Approach (2015, methodology): “There is a long history of using neural network for the task of face detection [33, 26, 7, 25].”
Toward automatic phenotyping of developing embryos from videos (2005, background): “Events can be global (for instance, which cells divide and in what order) or very specific (for instance, a visible featurecan appear between two specific cells at a specific time during the embryo development).”
Learning Speaker-Specific Characteristics With a Deep Neural Architecture (2011, background): “The principle behind unsupervised pretraining has been further investigated very recently [19].”
DeepFace: Closing the Gap to Human-Level Performance in Face Verification (2014, background): “Deep neural nets have also been applied in the past to face detection [24], face alignment [27] and face verification [8, 16].”