SF-Net: Single-Frame Supervision for Temporal Action Localization

When Does Self-Supervision Improve Few-Shot Learning?

The Geometry of Deep Generative Image Models

Is MLP-Mixer a CNN in Disguise?

Bringing Back MLPs

A Deep Dive into Neural Architecture Search Without Training (NASWOT)

Inject Noise to Remove Noise: A Deep Dive into Score-Based Generative Modeling Techniques

Generating Digital Painting Lighting Effects via RGB-space Geometry

Gym-μRTS: Toward Affordable Deep Reinforcement Learning Research in Real-Time Strategy Games

The Reality Behind the Optimization of Imaginary Variables

ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases

Funnel Activation for Visual Recognition (CVPR 2020)

Gated Channel Transformation for Visual Recognition (CVPR 2020)

Lyft's High-Capacity End-to-End Camera-Lidar Fusion for 3D Detection

Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training (ECCV 2020)

Sentiment Analysis with Bag of Tricks

Revisiting ResNets: Improved Training and Scaling Strategies

Explained: Characterizing Signal Propagation to Close the Performance Gap in Unnormalized ResNets

Strip Pooling: Rethinking Spatial Pooling for Scene Parsing (CVPR 2020)

Interventional Few-Shot Learning