Reproduce “VBot Exo2Ego Policy: Lightweight Data Collection and Exocentric-to-Egocentric Video Generation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tong Rao, Zhihao Zhang, Yuntian Wang, Siyu Zhou, Hailong Qin, Yinan Yu
VBot generates first-person video from third-person operation footage, using source video, text, and optionally an actor reference. Camera rays guide training but are absent at inference. On 546 validation clips, its first 49 frames score 15.856 dB PSNR and 191.55 FVD, versus EgoX’s 12.630 dB and 306.90. This comparison measures video fidelity, not robot performance or a uniquely true view.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.