Papers › Self-Supervised Model Adaptation for Multimodal Semantic Segmentation

Self-Supervised Model Adaptation for Multimodal Semantic Segmentation

11 Aug 2018arXiv:1808.03833archive 2025-07-28

Abhinav Valada, Rohit Mohan, Wolfram Burgard

Learning to reliably perceive and understand the scene is an integral enabler for robots to operate in the real-world. This problem is inherently challenging due to the multitude of object types as well as appearance changes caused by varying illumination and weather conditions. Leveraging complementary modalities can enable learning of semantically richer representations that are resilient to such perturbations. Despite the tremendous progress in recent years, most multimodal convolutional neural network approaches directly concatenate feature maps from individual modality streams rendering the model incapable of focusing only on relevant complementary information for fusion. To address this limitation, we propose a mutimodal semantic segmentation framework that dynamically adapts the fusion of modality-specific features while being sensitive to the object category, spatial location and scene context in a self-supervised manner. Specifically, we propose an architecture consisting of two modality-specific encoder streams that fuse intermediate encoder representations into a single decoder using our proposed self-supervised model adaptation fusion mechanism which optimally combines complementary features. As intermediate representations are not aligned across modalities, we introduce an attention scheme for better correlation. In addition, we propose a computationally efficient unimodal segmentation architecture termed AdapNet++ that incorporates a new encoder with multiscale residual units and an efficient atrous spatial pyramid pooling that has a larger effective receptive field with more than 10x fewer parameters, complemented with a strong decoder with a multi-resolution supervision scheme that recovers high-resolution details. Comprehensive empirical evaluations on several benchmarks demonstrate that both our unimodal and multimodal architectures achieve state-of-the-art performance.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

DeepSceneSeg/SSMA mentioned on GitHubtfGPL-3.0 report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

DecoderScene RecognitionSemantic Segmentationmodel

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Scene Recognition ScanNet SSMA Average Recall 54.28 #1 of 1 Archive leaderboard report
Semantic Segmentation Cityscapes test SSMA Mean IoU (class) 82.3% #29 of 105 Archive leaderboard report
Semantic Segmentation Cityscapes test AdapNet++ Mean IoU (class) 81.24% #46 of 105 Archive leaderboard report
Semantic Segmentation Freiburg Forest SSMA Mean IoU 84.18 #1 of 2 Archive leaderboard report
Semantic Segmentation Freiburg Forest AdapNet++ Mean IoU 83.09 #2 of 2 Archive leaderboard report
Semantic Segmentation SUN-RGBD TokenFusion (S) Mean IoU 45.73 #39 of 44 Archive leaderboard report
Semantic Segmentation SUN-RGBD DPLNet Mean IoU 38.4 #43 of 44 Archive leaderboard report
Semantic Segmentation SYNTHIA-CVPR’16 SSMA Mean IoU 92.1 #1 of 2 Archive leaderboard report
Semantic Segmentation SYNTHIA-CVPR’16 AdapNet++ Mean IoU 87.87 #2 of 2 Archive leaderboard report
Semantic Segmentation ScanNetV2 SSMA Mean IoU 57.7 #4 of 12 Archive leaderboard report
Semantic Segmentation ScanNetV2 AdapNet++ Mean IoU 50.3 #6 of 12 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Spatial Pyramid Pooling

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections