Boosting Self-Supervision for Single-View Scene Completion via Knowledge Distillation

arXiv:2404.07933v1 Announce Type: new
Abstract: Inferring scene geometry from images via Structure from Motion is a long-standing and fundamental problem in computer vision. While classical approaches and, more recently, depth map predictions only focus on the visible parts of a scene, the task of scene completion aims to reason about geometry even in occluded regions. With the popularity of neural radiance fields (NeRFs), implicit representations also became popular for scene completion by predicting so-called density fields. Unlike explicit approaches. e.g. …

abstract arxiv boosting computer computer vision cs.cv distillation focus geometry images knowledge map predictions reason scene geometry supervision type via view vision