1277 - A Variational Information Bottleneck Based Method to Compress Sequential Networks for Human

Highlights of Poster Session WED PM

Highlights from Poster Session TUE PM

Highlights of Poster Session TUE AM

Highlights from Poster Session THU PM

Highlight of Poster Session THU AM

Highlights of Poster Session WED AM

Clean-Label Backdoor Attacks on Video Recognition Models

Robust Superpixel-Guided Attentional Adversarial Attack

Look-Into-Object: Self-Supervised Structure Modeling for Object Recognition

Self-Robust 3D Point Recognition via Gather-Vector Guidance

In Defense of Grid Features for Visual Question Answering

Learning Temporal Co-Attention Models for Unsupervised Video Action Localization

Learning for Video Compression With Hierarchical Quality and Recurrent Enhancement

VIBE: Video Inference for Human Body Pose and Shape Estimation

Neural Blind Deconvolution Using Deep Priors

Adversarial Texture Optimization From RGB-D Scans

Universal Litmus Patterns: Revealing Backdoor Attacks in CNNs

8 - Joint Visual-Temporal Embedding for Unsupervised Learning of Actions in Untrimmed Sequences

33 - DeepOpht: Medical Report Generation for Retinal Images via Deep Models and Visual Explanation

232 - Improving Video Captioning with Temporal Composition of a Visual-Syntactic Embedding

743 - FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age for Bias Measurement and

1142 - Few-shot Font Style Transfer between Different Languages

740 - Efficient video annotation with visual interpolation and frame selection guidance

1315 - Data-free Knowledge Distillation for Object Detection