Understanding Multi-Granularity for Open-Vocabulary Part Segmentation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912146980339712 |
|---|---|
| author | Choi, Jiho Lee, Seonho Lee, Seungho Lee, Minhyun Shim, Hyunjung |
| author_facet | Choi, Jiho Lee, Seonho Lee, Seungho Lee, Minhyun Shim, Hyunjung |
| contents | Open-vocabulary part segmentation (OVPS) is an emerging research area focused on segmenting fine-grained entities using diverse and previously unseen vocabularies. Our study highlights the inherent complexities of part segmentation due to intricate boundaries and diverse granularity, reflecting the knowledge-based nature of part identification. To address these challenges, we propose PartCLIPSeg, a novel framework utilizing generalized parts and object-level contexts to mitigate the lack of generalization in fine-grained parts. PartCLIPSeg integrates competitive part relationships and attention control, alleviating ambiguous boundaries and underrepresented parts. Experimental results demonstrate that PartCLIPSeg outperforms existing state-of-the-art OVPS methods, offering refined segmentation and an advanced understanding of part relationships within images. Through extensive experiments, our model demonstrated a significant improvement over the state-of-the-art models on the Pascal-Part-116, ADE20K-Part-234, and PartImageNet datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_11384 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Understanding Multi-Granularity for Open-Vocabulary Part Segmentation Choi, Jiho Lee, Seonho Lee, Seungho Lee, Minhyun Shim, Hyunjung Computer Vision and Pattern Recognition Open-vocabulary part segmentation (OVPS) is an emerging research area focused on segmenting fine-grained entities using diverse and previously unseen vocabularies. Our study highlights the inherent complexities of part segmentation due to intricate boundaries and diverse granularity, reflecting the knowledge-based nature of part identification. To address these challenges, we propose PartCLIPSeg, a novel framework utilizing generalized parts and object-level contexts to mitigate the lack of generalization in fine-grained parts. PartCLIPSeg integrates competitive part relationships and attention control, alleviating ambiguous boundaries and underrepresented parts. Experimental results demonstrate that PartCLIPSeg outperforms existing state-of-the-art OVPS methods, offering refined segmentation and an advanced understanding of part relationships within images. Through extensive experiments, our model demonstrated a significant improvement over the state-of-the-art models on the Pascal-Part-116, ADE20K-Part-234, and PartImageNet datasets. |
| title | Understanding Multi-Granularity for Open-Vocabulary Part Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.11384 |