Enhancing 3D Semantic Scene Completion with a Refinement Module

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Dunxing, Lu, Jiachen, Yang, Han, Bao, Lei, Song, Bo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918517160280064
author Zhang, Dunxing
Lu, Jiachen
Yang, Han
Bao, Lei
Song, Bo
author_facet Zhang, Dunxing
Lu, Jiachen
Yang, Han
Bao, Lei
Song, Bo
contents We propose ESSC-RM, a plug-and-play Enhancing framework for Semantic Scene Completion with a Refinement Module, which can be seamlessly integrated into existing SSC models. ESSC-RM operates in two phases: a baseline SSC network first produces a coarse voxel prediction, which is subsequently refined by a 3D U-Net-based Prediction Noise-Aware Module (PNAM) and Voxel-level Local Geometry Module (VLGM) under multiscale supervision. Experiments on SemanticKITTI show that ESSC-RM consistently improves semantic prediction performance. When integrated into CGFormer and MonoScene, the mean IoU increases from 16.87% to 17.27% and from 11.08% to 11.51%, respectively. These results demonstrate that ESSC-RM serves as a general refinement framework applicable to a wide range of SSC models.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18363
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhancing 3D Semantic Scene Completion with a Refinement Module
Zhang, Dunxing
Lu, Jiachen
Yang, Han
Bao, Lei
Song, Bo
Computer Vision and Pattern Recognition
We propose ESSC-RM, a plug-and-play Enhancing framework for Semantic Scene Completion with a Refinement Module, which can be seamlessly integrated into existing SSC models. ESSC-RM operates in two phases: a baseline SSC network first produces a coarse voxel prediction, which is subsequently refined by a 3D U-Net-based Prediction Noise-Aware Module (PNAM) and Voxel-level Local Geometry Module (VLGM) under multiscale supervision. Experiments on SemanticKITTI show that ESSC-RM consistently improves semantic prediction performance. When integrated into CGFormer and MonoScene, the mean IoU increases from 16.87% to 17.27% and from 11.08% to 11.51%, respectively. These results demonstrate that ESSC-RM serves as a general refinement framework applicable to a wide range of SSC models.
title Enhancing 3D Semantic Scene Completion with a Refinement Module
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.18363