EgoAVFlow: Robot Policy Learning with Active Vision from Human Egocentric Videos via 3D Flow
Authors: Daesol Cho, Youngseok Jang, Danfei Xu, Sehoon Ha
Organizations: School of Interactive Computing, Georgia Institute of Technology, Georgia, USA · InnoCORE AI-Transformed Aerospace Research Center, KAIST, Daejeon, Republic of Korea
Egocentric human videos provide a scalable source of manipulation demonstrations; however, deploying them on robots requires active viewpoint control to maintain task-critical visibility, which human viewpoint imitation often fails to provide due to human-specific priors. We propose EgoAVFlow, which learns manipulation and active vision from egocentric videos through a shared 3D flow representation that supports geometric visibility reasoning and transfers without robot demonstrations. EgoAVFlow uses diffusion models to predict robot actions, future 3D flow, and camera trajectories, and refines viewpoints at test time with reward-maximizing denoising under a visibility-aware reward computed from predicted motion and scene geometry. Real-world experiments under actively changing viewpoints show that EgoAVFlow consistently outperforms prior human-demo-based baselines, demonstrating effective visibility maintenance and robust manipulation without robot demonstrations.
Figures & tables
Fig. 1 : EgoAVFlow learns manipulation and active viewpoint control from egocentric human videos by predicting future 3D flow and optimizing camera viewpoints for visibility, yielding viewpoint-robust robot execution without robot demonstrations.
Fig. 2 : Method overview. EgoAVFlow consists of three diffusion models. The robot policy πr produces future robot action sequences. The flow generation model f predicts future 3D flows from the outputs of πr . The view policy πv produces future camera viewpoints from the outputs of πr , f , and reconstructed mesh surfaces through a visibility-aware reward-maximizing denoising process. Viewpoints (A) represent that most query points are invisible ( Red LOS ) due to the table’s mesh surface or out of FoV, whereas in viewpoints (B) these points are visible ( Green LOS ), yielding a higher visibility reward.
Fig. 3 : Tasks. Each task requires appropriate viewpoint adjustments. Otherwise, the object is occluded by the robot or elements in the environment, such as a table or drawer.
Fig. 4 : Visibility comparison (best viewed in the digital version). The visibility is computed from each different fixed viewpoint. No single viewpoint can maintain full visibility throughout the execution, indicating that the viewpoint must be continuously adjusted online to maximize visibility.
Fig. 5 : Visibility reward. For all tasks, EgoAVFlow achieves higher average visibility rewards Rvis than HVI , demonstrating our method’s visibility maintenance capability. The error bars represent 1 standard error.
Fig. 6 : Qualitative comparison. Due to the visibility-maximizing viewpoint adjustments, EgoAVFlow maintains visibility of the query points and their predicted future flows, whereas HVI fails to keep them in view, causing the query points to move out of the FoV. All experimental figures and videos in Section V are best viewed in the supplementary video.
Method
Spray
Doll
Toilet Paper
Towel
V-B
HVI
5/25
10/25
7/25
5/25
V-C
AMPLIFY
4/25
6/25
9/25
9/25
EgoZero
10/25
7/25
7/25
9/25
Phantom
5/25
7/25
6/25
8/25
EgoAVFlow
20/25
18/25
17/25
18/25
( ×2.0 )
( ×2.5 )
( ×1.8 )
( ×2.0 )
TABLE I : Success rates of EgoAVFlow, HVI (Human Viewpoint Imitation), and robot policy learning baselines. All methods are trained on the same egocentric human video dataset (no robot data). Relative improvements compared to the second-best robot policy learning baseline are shown in parentheses .
Fig. 7 : Failure analysis. Top: Method-wise composition of failures for each category. Bottom: Example frames of a grasping failure. The breakdown suggests that manipulation-related failures in robot policy baselines are driven by a mismatch between their learned representations/assumptions, whereas human viewpoint imitation fails primarily due to the lack of visibility-aware viewpoint optimization.