Learning to generate diverse scene-aware and goal-oriented human motions in 3D scenes remains challenging due to the mediocre characters of the existing datasets on Human …
C Chen, U Jain, C Schissler, SVA Gari… - Computer Vision–ECCV …, 2020 - Springer
Moving around in the world is naturally a multisensory experience, but today's embodied agents are deaf—restricted to solely their visual perception of the environment. We introduce …
For robots to be generally useful, they must be able to find arbitrary objects described by people (ie, be language-driven) even without expensive navigation training on in-domain …
SM Park, YG Kim - Computer Science Review, 2023 - Elsevier
With the recent development of deep learning technology comes the wide use of artificial intelligence (AI) models in various domains. AI shows good performance for definite …
M Toromanoff, E Wirbel… - Proceedings of the IEEE …, 2020 - openaccess.thecvf.com
Reinforcement Learning (RL) aims at learning an optimal behavior policy from its own experiments and not rule-based control methods. However, there is no RL algorithm yet …
Does progress in simulation translate to progress on robots? If one method outperforms another in simulation, how likely is that trend to hold in reality on a robot? We examine this …
In reinforcement learning for visual navigation, it is common to develop a model for each new task, and train that model from scratch with task-specific interactions in 3D …
Navigation tasks in photorealistic 3D environments are challenging because they require perception and effective planning under partial observability. Recent work shows that map …
Can general-purpose neural models learn to navigate? For PointGoal navigation ("" go to x, y""), the answer is a clearyes'--mapless neural models composed of task-agnostic …