Published on:
30 April 2024
Primary Category:
Computer Vision and Pattern Recognition
Paper Authors:
Paul Engstler,
Andrea Vedaldi,
Iro Laina,
Christian Rupprecht
Proposes conditional depth completion model for scene generation
Self-supervised training scheme using teacher distillation
Evaluates depth predictions against ground truth geometry
State-of-the-art depth consistency on ScanNet and Hypersim datasets
Showcases approach by generating high-quality 360 degree scenes
Generating 3D Scenes with Depth Inpainting
This paper introduces two key innovations for generating 3D scenes from images. First, it develops a depth completion model to extrapolate missing depth values by conditioning on the existing scene geometry. This results in improved coherence compared to off-the-shelf depth estimators. Second, it provides a new benchmark to evaluate scene generation methods based on ground truth depth maps rather than image similarity metrics alone.
Virtual stereo matching for robust depth completion
Text-driven 3D scene synthesis
Depth: A Guide to Monocular Depth Estimation Using Self-Supervision
Generating 3D Shapes from Images via Multi-View RGB-D Modeling
Using simulations and AI for monocular depth data
Guiding unsupervised segmentation with depth and sampling
No comments yet, be the first to start the conversation...
Sign up to comment on this paper