Learning to Reconstruct 3D Non-Cuboid Room Layout from a Single RGB Image
Official implementation for Learning Multi-View Aggregation In the Wild for Large-Scale 3D Semantic Segmentation (CVPR'22 Best Paper Finalist 🎉) If you ❤️ or simply use this project, don't forget to give the repository a ⭐, it means a lot to us !
The 2D-3D-S dataset provides a variety of mutually registered modalities from 2D, 2.5D and 3D domains, with instance-level semantic and geometric annotations. It covers over 6,000 m2 collected in 6 large-scale indoor areas that originate from 3 different buildings. It contains over 70,000 RGB…
3RScan is a large scale, real-world dataset which features 1482 3D reconstructions / snapshots of 478 naturally changing indoor environments, designed for benchmarking emerging tasks such as long-term SLAM, scene change detection and object instance re-localization
waldjohannau/3rscan 的主要功能包括：Embodied Perception, Scene Understanding。
waldjohannau/3rscan 的开源替代品包括： baai-dcai/spatialbot. cyang0515/noncuboidroom — Learning to Reconstruct 3D Non-Cuboid Room Layout from a Single RGB Image. drprojects/deepviewagg — Official implementation for Learning Multi-View Aggregation In the Wild for Large-Scale 3D Semantic Segmentation… facebookresearch/habitat-api — Habitat-Lab. facebookresearch/habitat-sim — Habitat-sim is a high-performance 3D simulation platform designed for training and benchmarking embodied AI agents… alexsax/2d-3d-semantics — The 2D-3D-S dataset provides a variety of mutually registered modalities from 2D, 2.5D and 3D domains, with…