方法对比

	多模态语义分割	实例分割	语义分割	Vision Transformer
领域	深度学习	深度学习	深度学习	深度学习
方法族	Machine learning	Machine learning	Machine learning	Machine learning
起源年份	2014–2016	2017	2015	2021
提出者	Multiple contributors (Hazirbas et al., Long et al., and others)	He, K., Gkioxari, G., Dollar, P., Girshick, R.	Long, J., Shelhamer, E., & Darrell, T.	Dosovitskiy, A. et al.
类型	Pixel-level classification with multi-sensor fusion	Pixel-level detection and mask prediction	Dense prediction / pixel-wise classification	Transformer architecture for images (self-attention over patches)
开创性文献	Hazirbas, C., Ma, L., Domokos, C., & Cremers, D. (2016). FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-based CNN Architecture. In Proceedings of the Asian Conference on Computer Vision (ACCV). Springer.	He, K., Gkioxari, G., Dollar, P., & Girshick, R. (2017). Mask R-CNN. Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2961–2969.	Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3431–3440.	Dosovitskiy, A. et al. (2021). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. ICLR.
别名	multimodal scene parsing, multi-sensor semantic segmentation, RGB-D semantic segmentation, cross-modal semantic segmentation	instance-level segmentation, object instance segmentation, mask prediction, panoptic instance segmentation	pixel-wise classification, scene parsing, dense labeling, semantic scene segmentation	Görsel Transformer (ViT), görsel transformer, ViT, patch transformer for images
相关	3	4	5	5
摘要	Multimodal semantic segmentation assigns a semantic class label to every pixel in a scene by fusing information from two or more sensor modalities — most commonly RGB images paired with depth maps (RGB-D), LiDAR point clouds, thermal cameras, or text descriptions. Deep encoder-decoder networks learn to align and fuse complementary cues from each modality, producing denser and more accurate segmentation than any single-modality approach.	Instance segmentation is a computer vision task that simultaneously detects every distinct object in an image and produces a precise pixel-level mask for each individual object instance. Unlike semantic segmentation, which labels every pixel with a class, instance segmentation distinguishes between separate objects of the same class, enabling fine-grained spatial understanding.	Semantic segmentation assigns a class label to every pixel in an image, producing a dense, category-annotated map of the scene. Unlike object detection, which draws bounding boxes, it delineates the exact spatial extent of each class, making it indispensable in medical imaging, autonomous driving, satellite analysis, and any task where precise region boundaries matter.	The Vision Transformer (ViT), introduced by Dosovitskiy and colleagues in 2021, splits an image into fixed-size patches, treats those patches as a sequence, and applies the Transformer self-attention mechanism to image classification. Given enough training data, it surpasses convolutional neural networks (CNNs).