CSF-Net: Context-Semantic Fusion Network for Large Mask Inpainting
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909897957834752 |
|---|---|
| author | Heo, Chae-Yeon Cho, Yeong-Jun |
| author_facet | Heo, Chae-Yeon Cho, Yeong-Jun |
| contents | In this paper, we propose a semantic-guided framework to address the challenging problem of large-mask image inpainting, where essential visual content is missing and contextual cues are limited. To compensate for the limited context, we leverage a pretrained Amodal Completion (AC) model to generate structure-aware candidates that serve as semantic priors for the missing regions. We introduce Context-Semantic Fusion Network (CSF-Net), a transformer-based fusion framework that fuses these candidates with contextual features to produce a semantic guidance image for image inpainting. This guidance improves inpainting quality by promoting structural accuracy and semantic consistency. CSF-Net can be seamlessly integrated into existing inpainting models without architectural changes and consistently enhances performance across diverse masking conditions. Extensive experiments on the Places365 and COCOA datasets demonstrate that CSF-Net effectively reduces object hallucination while enhancing visual realism and semantic alignment. The code for CSF-Net is available at https://github.com/chaeyeonheo/CSF-Net. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_07987 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CSF-Net: Context-Semantic Fusion Network for Large Mask Inpainting Heo, Chae-Yeon Cho, Yeong-Jun Computer Vision and Pattern Recognition In this paper, we propose a semantic-guided framework to address the challenging problem of large-mask image inpainting, where essential visual content is missing and contextual cues are limited. To compensate for the limited context, we leverage a pretrained Amodal Completion (AC) model to generate structure-aware candidates that serve as semantic priors for the missing regions. We introduce Context-Semantic Fusion Network (CSF-Net), a transformer-based fusion framework that fuses these candidates with contextual features to produce a semantic guidance image for image inpainting. This guidance improves inpainting quality by promoting structural accuracy and semantic consistency. CSF-Net can be seamlessly integrated into existing inpainting models without architectural changes and consistently enhances performance across diverse masking conditions. Extensive experiments on the Places365 and COCOA datasets demonstrate that CSF-Net effectively reduces object hallucination while enhancing visual realism and semantic alignment. The code for CSF-Net is available at https://github.com/chaeyeonheo/CSF-Net. |
| title | CSF-Net: Context-Semantic Fusion Network for Large Mask Inpainting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.07987 |