# EgoWorld: Translating Exocentric View to Egocentric View using Rich Exocentric Observations Source: [EgoWorld: Translating Exocentric View to Egocentric View using Rich Exocentric Observations]() Collection index: [llms.txt]() Data updated: 2026-03-06T05:00:05.533Z Authors: Junho Park, Andrew Sangwoo Ye, Taein Kwon Published: 2025-06-22T04:21:48.000Z ## Abstract Egocentric vision is essential for both human and machine visual understanding, particularly in capturing the detailed hand-object interactions needed for manipulation tasks. Translating third-person views into first-person views significantly benefits augmented reality (AR), virtual reality (VR) and robotics applications. However, current exocentric-to-egocentric translation methods are limited by their dependence on 2D cues, synchronized multi-view settings, and unrealistic assumptions such as the necessity of an initial egocentric frame and relative camera poses during inference. To overcome these challenges, we introduce EgoWorld, a novel framework that reconstructs an egocentric view from rich exocentric observations, including point clouds, 3D hand poses, and textual descriptions. Our approach reconstructs a point cloud from estimated exocentric depth maps, reprojects it into the egocentric perspective, and then applies diffusion model to produce dense, semantically coherent egocentric images. Evaluated on four datasets (i.e., H2O, TACO, Assembly101, and Ego-Exo4D), EgoWorld achieves state-of-the-art performance and demonstrates robust generalization to new objects, actions, scenes, and subjects. Moreover, EgoWorld exhibits robustness on in-the-wild examples, underscoring its practical applicability. Project page is available at https://redorangeyellowy.github.io/EgoWorld/. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Exocentric-to-Egocentric Translation on TACO (unseen actions)](): Exocentric-to-Egocentric Translation; TACO (unseen actions); FID: 37.191. - [Exocentric-to-Egocentric Translation on Assembly101 (unseen actions)](): Exocentric-to-Egocentric Translation; Assembly101 (unseen actions); FID: 50.232. - [Exocentric-to-Egocentric Translation on Ego-Exo4D (unseen actions)](): Exocentric-to-Egocentric Translation; Ego-Exo4D (unseen actions); FID: 61.231. - [Exocentric-to-egocentric translation on H2O (unseen objects)](): Exocentric-to-Egocentric Translation; H2O (unseen objects); FID: 41.334. - [Exocentric-to-egocentric translation on H2O (unseen actions)](): Exocentric-to-Egocentric Translation; H2O (unseen actions); FID: 33.284. - [Exocentric-to-egocentric translation on H2O (unseen scenes)](): Exocentric-to-Egocentric Translation; H2O (unseen scenes); FID: 90.893. - [Exocentric-to-egocentric translation on H2O (unseen subjects)](): Exocentric-to-Egocentric Translation; H2O (unseen subjects); FID: 96.429. ## Original sources and implementation [Original paper]() [PDF]()