Current machine learning methods struggle to solve Bongard problems, whi...
We introduce PointOdyssey, a large-scale synthetic dataset, and data
gen...
We introduce TIDEE, an embodied agent that tidies up a disordered scene ...
Building 3D perception systems for autonomous vehicles that do not rely ...
Tracking pixels in videos is typically studied as an optical flow estima...
This paper explores self-supervised learning of amodal 3D feature
repres...
We propose an unsupervised method for detecting and tracking moving obje...
We present neural architectures that disentangle RGB-D images into objec...
We propose a system that learns to detect objects and infer their 3D pos...
We hypothesize that an agent that can look around in static scenes can l...
Consider the utterance "the tomato is to the left of the pot." Humans ca...
Humans can effortlessly imagine the occluded side of objects in a photog...
Cross-domain image-to-image translation should satisfy two requirements:...
Humans effortlessly "program" one another by communicating goals and des...
Researchers have developed excellent feed-forward models that learn to m...
Recently, convolutional networks (convnets) have proven useful for predi...
This paper proposes a new deep convolutional neural network (DCNN)
archi...
This paper presents a new state-of-the-art for document image classifica...