computer-vision-7

3D Semantic Scene Completion from a single RGB image

3260 papers • 126 benchmarks • 313 datasets

This task relies on a single RGB image to infer the dense 3D voxelized semantic scene.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in 3d-semantic-scene-completion-from-a-single-rgb-image-29

Trend

Dataset

Best Model

Actions

SemanticKITTI

NYUv2

KITTI-360

Libraries

i

Use these libraries to find 3d-semantic-scene-completion-from-a-single-rgb-image-29 models and implementations

Datasets

Subtasks

No subtasks available.

Most implemented papers

3D Sketch-Aware Semantic Scene Completion via Semi-Supervised Structure Prior

Xiaokang Chen, Hongsheng Li, Chen Qian, Kwan-Yee Lin, Gang Zeng•Mon Mar 30 2020

A new geometry-based strategy to embed depth information with low-resolution voxel representation, which could still be able to encode sufficient geometric information, e.g., room layout, object’s sizes and shapes, to infer the invisible areas of the scene with well structure-preserving details is proposed.

166

Content

0

Paper Graph

LMSCNet: Lightweight Multiscale 3D Semantic Completion

Raoul de Charette, Luis Roldão, Anne Verroust-Blondet•Sun Aug 23 2020

The ablation studies demonstrate the method is robust to lower density inputs, and that it enables very high speed semantic completion at the coarsest level, and provides a great performance/speed trade-off for mobile-robotics applications.

229 0

Paper Graph

Anisotropic Convolutional Networks for 3D Semantic Scene Completion

Peng Wang, Kai Han, J. Li, Yu Liu, Xia Yuan•Sat Apr 04 2020

A novel module called anisotropic convolution is proposed, which properties with flexibility and power impossible for the competing methods such as standard 3D convolution and some of its variations are proposed.

166 0

Paper Graph

Sparse Single Sweep LiDAR Point Cloud Segmentation via Learning Contextual Shape Priors from Scene Completion

Ruimao Zhang, Shuguang Cui, Jiantao Gao, Zhen Li, Xu Yan, Jie Li, Rui Huang•Sun Dec 06 2020

A novel sparse LiDAR point cloud semantic segmentation framework assisted by learned contextual shape priors is proposed, which inherently improves SS optimization through fully end-to-end training.

341 0

Paper Graph

MonoScene: Monocular 3D Semantic Scene Completion

Raoul de Charette, Anh-Quan Cao•Tue Nov 30 2021

Experiments show the MonoScene framework outperform the literature on all metries and datasets while hallucinating plausible scenery even beyond the camera field of view, and introduces a 3D context relation prior to enforce spatio-semantic consistency.

410 0

Paper Graph

VoxFormer: Sparse Voxel Transformer for Camera-Based 3D Semantic Scene Completion

Chaowei Xiao, S. Fidler, J. Álvarez, Chen Feng, Yiming Li, Zhiding Yu, Anima Anandkumar, C. Choy•Wed Feb 22 2023

VoxFormer, a Transformer-based semantic scene completion framework that can output complete 3D volumetric semantics from only 2D images that outperforms the state of the art with a relative improvement of 20.0% in geometry and 18.1% in semantics.

349 0

Paper Graph

OccFormer: Dual-path Transformer for Vision-based 3D Semantic Occupancy Prediction

Dalong Du, Yunpeng Zhang, Zhengbiao Zhu•Mon Apr 10 2023

OccFormer, a dual-path transformer network to effectively process the 3D volume for semantic occupancy prediction, achieves a long-range, dynamic, and efficient encoding of the camera-generated 3D voxel features.

309 0

Paper Graph

Symphonize 3D Semantic Scene Completion with Contextual Instance Queries

Tianheng Cheng, Xinggang Wang, Wenyu Liu, Hao Jiang, Naiyu Gao, Haoyang Zhang•Mon Jun 26 2023

A novel paradigm termed Symphonies (Scene-from-Insts), that delves into the integration of instance queries to orchestrate 2D-to-3D reconstruction and 3D scene modeling, and fosters holistic scene comprehension by capturing context through the efficient fusion of instance queries.

99 0

Paper Graph

NDC-Scene: Boost Monocular 3D Semantic Scene Completion in Normalized Device Coordinates Space

Hongsheng Li, Jiawei Yao, Chuming Li, Keqiang Sun, Yingjie Cai, Ouyang Wanli•Mon Sep 25 2023

A novel Normalized Device Coordinates scene completion network (NDC-Scene) is devised that directly extends the 2D feature map to a Normalized device coordinates space, rather than to the world space directly, through progressive restoration of the dimension of depth with deconvolution operations.

139 0

Paper Graph

Adding a benchmark result helps the community track progress.