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

Language-driven Semantic Segmentation

About

We present LSeg, a novel model for language-driven semantic image segmentation. LSeg uses a text encoder to compute embeddings of descriptive input labels (e.g., "grass" or "building") together with a transformer-based image encoder that computes dense per-pixel embeddings of the input image. The image encoder is trained with a contrastive objective to align pixel embeddings to the text embedding of the corresponding semantic class. The text embeddings provide a flexible label representation in which semantically similar labels map to similar regions in the embedding space (e.g., "cat" and "furry"). This allows LSeg to generalize to previously unseen categories at test time, without retraining or even requiring a single additional training sample. We demonstrate that our approach achieves highly competitive zero-shot performance compared to existing zero- and few-shot semantic segmentation methods, and even matches the accuracy of traditional segmentation algorithms when a fixed label set is provided. Code and demo are available at https://github.com/isl-org/lang-seg.

Boyi Li, Kilian Q. Weinberger, Serge Belongie, Vladlen Koltun, Ren\'e Ranftl• 2022

Related benchmarks

TaskDatasetResultRank
Semantic segmentationADE20K (val)
mIoU47.25
2731
Semantic segmentationADE20K
mIoU18
936
Semantic segmentationPASCAL VOC (val)
mIoU59
338
Semantic segmentationPascal VOC (test)
mIoU47.4
236
Semantic segmentationScanNet (val)
mIoU47.5
231
Semantic segmentationADE20K A-150
mIoU18
188
Few-shot Semantic SegmentationPASCAL-5^i (test)
FB-IoU67
177
Semantic segmentationPascal Context 59
mIoU46.5
164
3D Visual GroundingScanRefer (val)--
155
Semantic segmentationPASCAL-Context 59 class (val)
mIoU46.5
125
Showing 10 of 89 rows
...

Other info

Code

Follow for update