CASSC: Context‐aware method for depth guided semantic scene completion

Abstract Semantic scene completion is a crucial end‐to‐end 3D perception task, and the 3D information perception subjects is vital for autonomous driving. This paper presents CASSC, a novel adaptive context‐aware method based on Transformer networks, aimed at realizing camera‐based semantic scene co...

Full description

Saved in:
Bibliographic Details
Main Authors: Jinghao Cao, Ming Li, Sheng Liu, Yang Li, Sidan Du
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:IET Image Processing
Subjects:
Online Access:https://doi.org/10.1049/ipr2.13280
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850118349429669888
author Jinghao Cao
Ming Li
Sheng Liu
Yang Li
Sidan Du
author_facet Jinghao Cao
Ming Li
Sheng Liu
Yang Li
Sidan Du
author_sort Jinghao Cao
collection DOAJ
description Abstract Semantic scene completion is a crucial end‐to‐end 3D perception task, and the 3D information perception subjects is vital for autonomous driving. This paper presents CASSC, a novel adaptive context‐aware method based on Transformer networks, aimed at realizing camera‐based semantic scene completion algorithms. The key idea is to leverage rich context information from images to obtain pixel‐level label proposals, followed by designing a multiscale fusion mechanism to merge this information and match it with voxel space. A weakly supervised training strategy is proposed to obtain semantic label distribution features from images and introduce an adaptive multiscale fusion module to fuse and adaptively match these features with voxel space. Here, CASSC achieves state‐of‐the‐art performance on the SemanticKITTI dataset and demonstrates excellent performance on the SSC‐Bench dataset. Ablation experiments validate the rationality and effectiveness of our design, and the model and code of CASSC will be open‐sourced on https://github.com/dogooooo/CASSC.
format Article
id doaj-art-c7baabe01d664fd0888a9a4456b5ecbf
institution OA Journals
issn 1751-9659
1751-9667
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series IET Image Processing
spelling doaj-art-c7baabe01d664fd0888a9a4456b5ecbf2025-08-20T02:35:53ZengWileyIET Image Processing1751-96591751-96672024-12-0118144716473010.1049/ipr2.13280CASSC: Context‐aware method for depth guided semantic scene completionJinghao Cao0Ming Li1Sheng Liu2Yang Li3Sidan Du4School of Electronic Science and Engineering Nanjing University Nanjing Jiangsu ChinaSchool of Electronic Science and Engineering Nanjing University Nanjing Jiangsu ChinaSchool of Electronic Science and Engineering Nanjing University Nanjing Jiangsu ChinaSchool of Electronic Science and Engineering Nanjing University Nanjing Jiangsu ChinaSchool of Electronic Science and Engineering Nanjing University Nanjing Jiangsu ChinaAbstract Semantic scene completion is a crucial end‐to‐end 3D perception task, and the 3D information perception subjects is vital for autonomous driving. This paper presents CASSC, a novel adaptive context‐aware method based on Transformer networks, aimed at realizing camera‐based semantic scene completion algorithms. The key idea is to leverage rich context information from images to obtain pixel‐level label proposals, followed by designing a multiscale fusion mechanism to merge this information and match it with voxel space. A weakly supervised training strategy is proposed to obtain semantic label distribution features from images and introduce an adaptive multiscale fusion module to fuse and adaptively match these features with voxel space. Here, CASSC achieves state‐of‐the‐art performance on the SemanticKITTI dataset and demonstrates excellent performance on the SSC‐Bench dataset. Ablation experiments validate the rationality and effectiveness of our design, and the model and code of CASSC will be open‐sourced on https://github.com/dogooooo/CASSC.https://doi.org/10.1049/ipr2.13280computer visionconvolutional neural netsimage processingsupervised learning
spellingShingle Jinghao Cao
Ming Li
Sheng Liu
Yang Li
Sidan Du
CASSC: Context‐aware method for depth guided semantic scene completion
IET Image Processing
computer vision
convolutional neural nets
image processing
supervised learning
title CASSC: Context‐aware method for depth guided semantic scene completion
title_full CASSC: Context‐aware method for depth guided semantic scene completion
title_fullStr CASSC: Context‐aware method for depth guided semantic scene completion
title_full_unstemmed CASSC: Context‐aware method for depth guided semantic scene completion
title_short CASSC: Context‐aware method for depth guided semantic scene completion
title_sort cassc context aware method for depth guided semantic scene completion
topic computer vision
convolutional neural nets
image processing
supervised learning
url https://doi.org/10.1049/ipr2.13280
work_keys_str_mv AT jinghaocao cassccontextawaremethodfordepthguidedsemanticscenecompletion
AT mingli cassccontextawaremethodfordepthguidedsemanticscenecompletion
AT shengliu cassccontextawaremethodfordepthguidedsemanticscenecompletion
AT yangli cassccontextawaremethodfordepthguidedsemanticscenecompletion
AT sidandu cassccontextawaremethodfordepthguidedsemanticscenecompletion