Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Efficient Multimodal Semantic Segmentation via Dual-Prompt Learning

About

Multimodal (e.g., RGB-Depth/RGB-Thermal) fusion has shown great potential for improving semantic segmentation in complex scenes (e.g., indoor/low-light conditions). Existing approaches often fully fine-tune a dual-branch encoder-decoder framework with a complicated feature fusion strategy for achieving multimodal semantic segmentation, which is training-costly due to the massive parameter updates in feature extraction and fusion. To address this issue, we propose a surprisingly simple yet effective dual-prompt learning network (dubbed DPLNet) for training-efficient multimodal (e.g., RGB-D/T) semantic segmentation. The core of DPLNet is to directly adapt a frozen pre-trained RGB model to multimodal semantic segmentation, reducing parameter updates. For this purpose, we present two prompt learning modules, comprising multimodal prompt generator (MPG) and multimodal feature adapter (MFA). MPG works to fuse the features from different modalities in a compact manner and is inserted from shadow to deep stages to generate the multi-level multimodal prompts that are injected into the frozen backbone, while MPG adapts prompted multimodal features in the frozen backbone for better multimodal semantic segmentation. Since both the MPG and MFA are lightweight, only a few trainable parameters (3.88M, 4.4% of the pre-trained backbone parameters) are introduced for multimodal feature fusion and learning. Using a simple decoder (3.27M parameters), DPLNet achieves new state-of-the-art performance or is on a par with other complex approaches on four RGB-D/T semantic segmentation datasets while satisfying parameter efficiency. Moreover, we show that DPLNet is general and applicable to other multimodal tasks such as salient object detection and video semantic segmentation. Without special design, DPLNet outperforms many complicated models. Our code will be available at github.com/ShaohuaDong2021/DPLNet.

Shaohua Dong, Yunhe Feng, Qing Yang, Yan Huang, Dongfang Liu, Heng Fan• 2023

Related benchmarks

TaskDatasetResultRank
Semantic segmentationSUN RGB-D (test)
mIoU52.8
191
Semantic segmentationNYUD v2 (test)
mIoU59.3
187
RGB-D Salient Object DetectionSIP
S-measure (Sα)0.89
124
RGB-D Salient Object DetectionLFSD
S-measure (Sα)87.3
122
RGBD Saliency DetectionDES
S-measure0.94
102
Semantic segmentationNYU V2
mIoU59.3
74
Semantic segmentationPST900 (test)
mIoU86.7
72
RGB-D Saliency DetectionNLPR
Max F-beta0.897
65
Salient Object DetectionVT5000
S-Measure0.879
50
Salient Object DetectionNJU2K
S-measure (Sα)92
23
Showing 10 of 14 rows

Other info

Code

Follow for update