Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2509.19589 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916966795575296 |
|---|---|
| author | Menn, Dennis Liang, Feng Marculescu, Diana |
| author_facet | Menn, Dennis Liang, Feng Marculescu, Diana |
| contents | Artifact detectors have been shown to enhance the performance of image-generative models by serving as reward models during fine-tuning. These detectors enable the generative model to improve overall output fidelity and aesthetics. However, training the artifact detector requires expensive pixel-level human annotations that specify the artifact regions. The lack of annotated data limits the performance of the artifact detector. A naive pseudo-labeling approach-training a weak detector and using it to annotate unlabeled images-suffers from noisy labels, resulting in poor performance. To address this, we propose an artifact corruption pipeline that automatically injects artifacts into clean, high-quality synthetic images on a predetermined region, thereby producing pixel-level annotations without manual labeling. The proposed method enables training of an artifact detector that achieves performance improvements of 13.2% for ConvNeXt and 3.7% for Swin-T, as verified on human-labeled data, compared to baseline approaches. This work represents an initial step toward scalable pixel-level artifact annotation datasets that integrate world knowledge into artifact detection. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_19589 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Synthesizing Artifact Dataset for Pixel-level Detection Menn, Dennis Liang, Feng Marculescu, Diana Computer Vision and Pattern Recognition Artifact detectors have been shown to enhance the performance of image-generative models by serving as reward models during fine-tuning. These detectors enable the generative model to improve overall output fidelity and aesthetics. However, training the artifact detector requires expensive pixel-level human annotations that specify the artifact regions. The lack of annotated data limits the performance of the artifact detector. A naive pseudo-labeling approach-training a weak detector and using it to annotate unlabeled images-suffers from noisy labels, resulting in poor performance. To address this, we propose an artifact corruption pipeline that automatically injects artifacts into clean, high-quality synthetic images on a predetermined region, thereby producing pixel-level annotations without manual labeling. The proposed method enables training of an artifact detector that achieves performance improvements of 13.2% for ConvNeXt and 3.7% for Swin-T, as verified on human-labeled data, compared to baseline approaches. This work represents an initial step toward scalable pixel-level artifact annotation datasets that integrate world knowledge into artifact detection. |
| title | Synthesizing Artifact Dataset for Pixel-level Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2509.19589 |