ObjEmbed-2B

ObjEmbed: Towards Universal Multimodal Object Embeddings

ObjEmbed is a multimodal embedding model designed to align specific image regions (objects) with textual descriptions. Unlike global embedding models, ObjEmbed decomposes an image into multiple regional embeddings along with global embeddings, supporting tasks such as visual grounding, local image retrieval, and global image retrieval.

Key Features

• Object-Oriented Representation: It captures both semantic and spatial aspects of objects by generating two complementary embeddings for each region: an object embedding for semantic matching and an IoU embedding that predicts localization quality.
• Versatility: It seamlessly handles both region-level and image-level tasks.
• Efficient Encoding: All objects in an image, along with the full image, are encoded in a single forward pass for high efficiency.

Resources

• Paper: ObjEmbed: Towards Universal Multimodal Object Embeddings
• Code: Official GitHub Repository

Citation

If you find ObjEmbed helpful for your research, please consider citing:

@article{fu2026objembed,
  title={ObjEmbed: Towards Universal Multimodal Object Embeddings},
  author={Fu, Shenghao and Su, Yukun and Rao, Fengyun and LYU, Jing and Xie, Xiaohua and Zheng, Wei-Shi},
  journal={arXiv preprint arXiv:2602.01753},
  year={2026}
}