Modeling local and global deformations in Deep Learning: Epitomic convolution, Multiple Instance Learning, and sliding window detection

Papandreou, George; Kokkinos, Iasonas; Savalle, Pierre-André

doi:10.1109/cvpr.2015.7298636

Cited by 198 publications

(149 citation statements)

References 19 publications

Supporting

Mentioning

147

Contrasting

Unclassified

Order By: Relevance

“…This is achieved by stochastic gradient descent training using the network architecture and cost function described in Section 3, which explicitly searches for the best candidate position of the object in the image using the global max-pooling operation. We also search over object scales (similar to [40]) by training from images of different sizes. The training procedure is illustrated in Figure 2.…”

Section: Weakly Supervised Learning and Classificationmentioning

confidence: 99%

Is object localization for free? - Weakly-supervised learning with convolutional neural networks

Oquab

Bottou

Laptev

et al. 2015

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

811

756

View full text Add to dashboard Cite

Successful methods for visual object recognition typically rely on training datasets containing lots of richly annotated images. Detailed image annotation, e.g. by object bounding boxes, however, is both expensive and often subjective. We describe a weakly supervised convolutional neural network (CNN) for object classification that relies only on image-level labels, yet can learn from cluttered scenes containing multiple objects. We quantify its object classification and object location prediction performance on the Pascal VOC 2012 (20 object classes) and the much larger Microsoft COCO (80 object classes) datasets. We find that the network (i) outputs accurate image-level labels, (ii) predicts approximate locations (but not extents) of objects, and (iii) performs comparably to its fully-supervised counterparts using object bounding box annotation for training.

show abstract

Section: Weakly Supervised Learning and Classificationmentioning

confidence: 99%

Is object localization for free? - Weakly-supervised learning with convolutional neural networks

Oquab

Bottou

Laptev

et al. 2015

2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

811

756

View full text Add to dashboard Cite

show abstract

“…We first filter left image and reconstruction error and left disparity and geometric error map E g independently by using one layer of convolution followed by batch normalization. Both these results are then concatenated and followed by atrous convolution [18] to sample from a larger context without increasing the network size. We used dilations with rate 1, 2, 4, 8, 1, and 1 respectively.…”

Section: Disparity Refinementmentioning

confidence: 99%

StereoDRNet: Dilated Residual StereoNet

Chabra

Straub

Sweeney

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Figure 1: StereoDRNet enables estimation of high quality depth maps that opens the door to high quality reconstruction by passive stereo video. In this figure we compare the output from dense reconstruction [15] built form depth maps generated by StereoDRNet, PSMNet [2] and a structured light system [25] (termed Ground Truth). We report and visualize point-toplane distance RMS error on the reconstructed meshes with respect to the ground truth demonstrating the improvement in reconstruction over the state-of-the-art. AbstractWe propose a system that uses a convolution neural network (CNN) to estimate depth from a stereo pair followed by volumetric fusion of the predicted depth maps to produce a 3D reconstruction of a scene. Our proposed depth refinement architecture, predicts view-consistent disparity and occlusion maps that helps the fusion system to produce geometrically consistent reconstructions. We utilize 3D dilated convolutions in our proposed cost filtering network that yields better filtering while almost halving the computational cost in comparison to state of the art cost filtering architectures. For feature extraction we use the Vortex Pooling architecture [26]. The proposed method achieves state of the art results in KITTI 2012, KITTI 2015 and ETH 3D stereo benchmarks. Finally, we demonstrate that our system is able to produce high fidelity 3D scene reconstructions that outperforms the state of the art stereo system.

show abstract

“…We appropriately place interpolation layers to ensure that results from different skip layers have commensurate dimensions, while, as in [40,10], we use atrous convolution to increase the spatial resolution of high-level neurons. Finally, to account for the varying face sizes in images we employ a 3-scale pyramid of our proposed network where at scales 2 & 3 we down-sample the image by half and a quarter times respectively by using a 2D average pooling operation, similar to [33].…”

Section: Modelmentioning

confidence: 99%

Face Normals "In-the-Wild" Using Fully Convolutional Networks

Trigeorgis

Snape

Kokkinos

et al. 2017

2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)

Self Cite

View full text Add to dashboard Cite

In this work we pursue a data-driven approach to the problem of estimating surface normals from a single intensity image, focusing in particular on human faces. We introduce new methods to exploit the currently available facial databases for dataset construction and tailor a deep convolutional neural network to the task of estimating facial surface normals 'in-the-wild'. We train a fully convolutional network that can accurately recover facial normals from images including a challenging variety of expressions and facial poses. We compare against state-of-the-art face Shape-from-Shading and 3D reconstruction techniques and show that the proposed network can recover substantially more accurate and realistic normals. Furthermore, in contrast to other existing face-specific surface recovery methods, we do not require the solving of an explicit alignment step due to the fully convolutional nature of our network.

show abstract

Modeling local and global deformations in Deep Learning: Epitomic convolution, Multiple Instance Learning, and sliding window detection

Cited by 198 publications

References 19 publications

Is object localization for free? - Weakly-supervised learning with convolutional neural networks

Is object localization for free? - Weakly-supervised learning with convolutional neural networks

StereoDRNet: Dilated Residual StereoNet

Face Normals "In-the-Wild" Using Fully Convolutional Networks

Contact Info

Product

Resources

About