RefVOS: A Closer Look at Referring Expressions for Video Object Segmentation
About
The task of video object segmentation with referring expressions (language-guided VOS) is to, given a linguistic phrase and a video, generate binary masks for the object to which the phrase refers. Our work argues that existing benchmarks used for this task are mainly composed of trivial cases, in which referents can be identified with simple phrases. Our analysis relies on a new categorization of the phrases in the DAVIS-2017 and Actor-Action datasets into trivial and non-trivial REs, with the non-trivial REs annotated with seven RE semantic categories. We leverage this data to analyze the results of RefVOS, a novel neural network that obtains competitive results for the task of language-guided image segmentation and state of the art results for language-guided VOS. Our study indicates that the major challenges for the task are related to understanding motion and static actions.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Referring Image Segmentation | RefCOCO+ (test-B) | mIoU36.17 | 200 | |
| Referring Image Segmentation | RefCOCO (val) | -- | 197 | |
| Referring Image Segmentation | RefCOCO (test A) | -- | 178 | |
| Video segmentation from a sentence | A2D Sentences (test) | Overall IoU67.2 | 122 | |
| Referring Image Segmentation | RefCOCO (test-B) | -- | 119 | |
| Referring Image Segmentation | RefCOCO+ (val) | -- | 117 | |
| Referring Image Segmentation | RefCOCO+ (test-A) | -- | 89 | |
| Language-guided Video Object Segmentation | DAVIS 1st frame Referring Expressions 2017 (val) | J&F Score44.5 | 6 | |
| Language-guided Video Object Segmentation | A2D (test) | Precision @0.557.8 | 5 | |
| Language-guided Video Object Segmentation | DAVIS full video Referring Expressions 2017 (val) | J&F Score45.1 | 5 |