Computer Vision – ECCV 2016 Workshops: Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part III

Front Matter....Pages I-XXIII
Front Matter....Pages 1-2
Back to Basics: Unsupervised Learning of Optical Flow via Brightness Constancy and Motion Smoothness....Pages 3-10
Human Action Recognition Without Human....Pages 11-17
Motion Representation with Acceleration Images....Pages 18-24
Segmentation Free Object Discovery in Video....Pages 25-31
Human Pose Estimation in Space and Time Using 3D CNN....Pages 32-39
Autonomous Driving Challenge: To Infer the Property of a Dynamic Object Based on Its Motion Pattern....Pages 40-46
Temporal Convolutional Networks: A Unified Approach to Action Segmentation....Pages 47-54
Making a Case for Learning Motion Representations with Phase....Pages 55-64
Front Matter....Pages 65-66
gvnn: Neural Network Library for Geometric Computer Vision....Pages 67-82
On-Line Large Scale Semantic Fusion....Pages 83-99
Learning Covariant Feature Detectors....Pages 100-117
Scene Segmentation Driven by Deep Learning and Surface Fitting....Pages 118-132
Improving Constrained Bundle Adjustment Through Semantic Scene Labeling....Pages 133-142
A CNN Cascade for Landmark Guided Semantic Part Segmentation....Pages 143-155
3D Human Pose Estimation Using Convolutional Neural Networks with 2D Pose Information....Pages 156-169
Overcoming Occlusion with Inverse Graphics....Pages 170-185
Deep Kinematic Pose Regression....Pages 186-201
How Useful Is Photo-Realistic Rendering for Visual Learning?....Pages 202-217
Learning the Structure of Objects from Web Supervision....Pages 218-235
VConv-DAE: Deep Volumetric Shape Learning Without Object Labels....Pages 236-250
Front Matter....Pages 65-66
Deep Shape from a Low Number of Silhouettes....Pages 251-265
Deep Disentangled Representations for Volumetric Reconstruction....Pages 266-279
Class-Specific Object Pose Estimation and Reconstruction Using 3D Part Geometry....Pages 280-295
Monocular Surface Reconstruction Using 3D Deformable Part Models....Pages 296-308
Front Matter....Pages 309-310
Deep Bimodal Regression for Apparent Personality Analysis....Pages 311-324
SASE: RGB-Depth Database for Human Head Pose Estimation....Pages 325-336
Bi-modal First Impressions Recognition Using Temporally Ordered Deep Audio and Stochastic Visual Features....Pages 337-348
Deep Impression: Audiovisual Deep Residual Networks for Multimodal Apparent Personality Trait Recognition....Pages 349-358
Deep Learning for Facial Action Unit Detection Under Large Head Poses....Pages 359-371
Combining Deep Facial and Ambient Features for First Impression Estimation....Pages 372-385
The Static Multimodal Dyadic Behavior Dataset for Engagement Prediction....Pages 386-399
ChaLearn LAP 2016: First Round Challenge on First Impressions - Dataset and Results....Pages 400-418
Overcoming Calibration Problems in Pattern Labeling with Pairwise Ratings: Application to Personality Traits....Pages 419-432
Front Matter....Pages 433-434
Best Practices for Fine-Tuning Visual Classifiers to New Domains....Pages 435-442
Deep CORAL: Correlation Alignment for Deep Domain Adaptation....Pages 443-450
Nonlinear Embedding Transform for Unsupervised Domain Adaptation....Pages 451-457
Unsupervised Domain Adaptation with Regularized Domain Instance Denoising....Pages 458-466
Online Heterogeneous Transfer Learning by Weighted Offline and Online Classifiers....Pages 467-474
Learning the Roots of Visual Domain Shift....Pages 475-482
Heterogeneous Face Recognition with CNNs....Pages 483-491
Front Matter....Pages 433-434
VLAD Is not Necessary for CNN....Pages 492-499
Training a Mentee Network by Transferring Knowledge from a Mentor Network....Pages 500-507
Lightweight Unsupervised Domain Adaptation by Convolutional Filter Reconstruction....Pages 508-515
Deep Attributes for One-Shot Face Recognition....Pages 516-523
Hard Negative Mining for Metric Learning Based Zero-Shot Classification....Pages 524-531
Transfer Learning for Cell Nuclei Classification in Histopathology Images....Pages 532-539
Visual Analogies: A Framework for Defining Aspect Categorization....Pages 540-547
Front Matter....Pages 549-550
A Direct Method for Robust Model-Based 3D Object Tracking from a Monocular RGB Image....Pages 551-562
A Radial Search Method for Fast Nearest Neighbor Search on Range Images....Pages 563-577
Direct-from-Video: Unsupervised NRSfM....Pages 578-594
Physical Reasoning for 3D Object Recognition Using Global Hypothesis Verification....Pages 595-605
On Evaluation of 6D Object Pose Estimation....Pages 606-619
Reconstructing Articulated Rigged Models from RGB-D Videos....Pages 620-633
RobotFusion: Grasping with a Robotic Manipulator via Multi-view Reconstruction....Pages 634-647
SemanticFusion: Joint Labeling, Tracking and Mapping....Pages 648-664
Towards Categorization and Pose Estimation of Sets of Occluded Objects in Cluttered Scenes from Depth Data and Generic Object Models Using Joint Parsing....Pages 665-681
Front Matter....Pages 683-684
The CUDA LATCH Binary Descriptor: Because Sometimes Faster Means Better....Pages 685-697
Sensor Fusion for Sparse SLAM with Descriptor Pooling....Pages 698-710
An Evaluation of Local Feature Detectors and Descriptors for Infrared Images....Pages 711-723
Evaluating Local Features for Day-Night Matching....Pages 724-736
Front Matter....Pages 683-684
Image Correspondences Matching Using Multiple Features Fusion....Pages 737-746
Learning Local Convolutional Features for Face Recognition with 2D-Warping....Pages 747-758
Improving Performances of MSER Features in Matching and Retrieval Tasks....Pages 759-770
Front Matter....Pages 771-772
Improved Image Boundaries for Better Video Segmentation....Pages 773-788
Point-Wise Mutual Information-Based Video Segmentation with High Temporal Consistency....Pages 789-803
Can Ground Truth Label Propagation from Video Help Semantic Segmentation?....Pages 804-820
3D Point Cloud Video Segmentation Based on Interaction Analysis....Pages 821-835
Deep Learning Based Fence Segmentation and Removal from an Image Using a Video Sequence....Pages 836-851
Clockwork Convnets for Video Semantic Segmentation....Pages 852-868
Front Matter....Pages 869-870
Learning Markerless Human Pose Estimation from Multiple Viewpoint Video....Pages 871-878
Virtual Immortality: Reanimating Characters from TV Shows....Pages 879-886
Light Source Estimation in Synthetic Images....Pages 887-893
LEE: A Photorealistic Virtual Environment for Assessing Driver-Vehicle Interactions in Self-driving Mode....Pages 894-900
Enhancing Place Recognition Using Joint Intensity - Depth Analysis and Synthetic Data....Pages 901-908
UnrealCV: Connecting Computer Vision to Unreal Engine....Pages 909-916
Back Matter....Pages 917-919