Self-Supervised Model Adaptation for Multimodal Semantic Segmentation

11 Aug 2018  ·  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.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Semantic Segmentation Cityscapes test AdapNet++ Mean IoU (class) 81.24% # 41
Semantic Segmentation Cityscapes test SSMA Mean IoU (class) 82.3% # 25
Semantic Segmentation Freiburg Forest SSMA Mean IoU 84.18 # 1
Semantic Segmentation Freiburg Forest AdapNet++ Mean IoU 83.09 # 2
Scene Recognition ScanNet SSMA Average Recall 54.28 # 1
Semantic Segmentation ScanNetV2 SSMA Mean IoU 57.7 # 3
Semantic Segmentation ScanNetV2 AdapNet++ Mean IoU 50.3 # 5
Semantic Segmentation SUN-RGBD AdapNet++ Mean IoU 38.4 # 33
Semantic Segmentation SUN-RGBD SSMA Mean IoU 45.73 # 29
Semantic Segmentation SYNTHIA-CVPR’16 SSMA Mean IoU 92.1 # 1
Semantic Segmentation SYNTHIA-CVPR’16 AdapNet++ Mean IoU 87.87 # 2