Detecting Invisible People

Human-to-Robot Imitation in the Wild

Fuzzy Metaballs: Approximate Differentiable Rendering with Algebraic Surfaces

Dataset Distillation by Matching Training Trajectories

RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wild

Efficient Spatially Sparse Inference for Conditional GANs and Diffusion Models

Compositional and Scalable Object SLAM

Shelf-Supervised Mesh Prediction in the Wild

Dynamics-Regulated Kinematic Policy for Egocentric Pose Estimation

Dual-Shutter Optical Vibration Sensing

Ensembling Off-the-shelf Models for GAN Training

BANMo: Building Animatable 3D Neural Models from Many Casual Videos

Bottom Up Top Down Detection Transformers for Language Grounding in Images and Point Clouds

Differentiable Raycasting for Self-supervised Occupancy Forecasting

Sequential Voting with Relational Box Fields for Active Object Detection