3D Gaussian Splatting for Real-Time Radiance Field Rendering
Proposition of a new method for real-time rendering of radiance fields that achieves great visual quality using a 3D Gaussian scene representation.
Read here
Learning Robust Perceptive Locomotion for Quadrupedal Robots in the Wild
Presenting a method for combining exteroceptive and proprioceptive perception for legged locomotion control, achieving high robustness and speed in a variety of challenging environments.
Read here
Reinforcement Learning, Fast and Slow
Recent advances in deep reinforcement learning have made it more efficient and potentially more relevant to understanding human learning. Should we let AI model the human brain?
Read here