← Search

Chengmin Gao

2 accepted papers

2024

Improving Viewpoint-Independent Object-Centric Representations through Active Viewpoint Selection

NeurIPS 2024poster

Given the complexities inherent in visual scenes, such as object occlusion, a comprehensive understanding often requires observation from multiple viewpoints. Existing multi-viewpoint object-centric learning methods typically employ random or sequential viewpoint selection strategies. While applicab…

Cited by 0SourcePDFScholar
2023

Time-Conditioned Generative Modeling of Object-Centric Representations for Video Decomposition and Prediction

UAI 2023poster

When perceiving the world from multiple viewpoints, humans have the ability to reason about the complete objects in a compositional manner even when an object is completely occluded from certain viewpoints. Meanwhile, humans are able to imagine novel views after observing multiple viewpoints. Recent…