Bbox Indicator Visual Baseline Vi
This repository contains the visual-only baseline using the target-region bbox indicator input path.
Source du modèle
- Type
- Other
- Taille du fichier
- 3,49 GB
- Version
- vModel state
- Mis à jour
- 10 oct. 2026
- Source
- Hugging Face
Description de la source
BBox-Indicator Visual Baseline (Vietnamese)
This repository contains the visual-only baseline using the target-region bbox indicator input path.
Architecture
full image + target region bbox [cx, cy, w, h] / 1000
-> SigLIP2 patch tokens
-> Linear(4, 1024) bbox indicator token
-> prepend bbox indicator before patch tokens
-> semantic projector
-> mBART decoder with LoRA
-> Vietnamese region caption
Main files
model_state.pt
model_config.json
tokenizer/
image_processor/
bbox_indicator_baseline/
requirements.txt
Sources
1 source
Sources
1 sourceVérifié 10 oct.
Artefacts du modèle
1 artefact
Artefacts du modèle
1 artefactExtraits de sources
2 extraits
Extraits de sources
2 extraits--- language: - vi pipeline_tag: image-to-text tags: - image-captioning - siglip2 - mbart - lora - bbox-indicator - visual-genome license: mit --- # BBox-Indicator Visual Baseline (Vietnamese) This repository contains the visual-only baseline using the target-region bbox indicator input path. ## Architecture ```text full image + target region bbox [cx, cy, w, h] / 1000 -> SigLIP2 patch tokens -> Linear(4, 1024) bbox indicator token -> prepend bbox indicator before patch tokens -> semantic projector -> mBART decoder with LoRA -> Vietnamese region caption ``` ## Main files ```text model_state.pt model_config.json tokenizer/ image_processor/ bbox_indicator_baseline/ requirements.txt ```
Source context: 0 downloads · 0 likes · Pipeline image-to-text · Repo AnhDau/bbox-indicator-visual-baseline-vi
