CoDet: Co-Occurrence Guided Region-Word Alignment for Open-Vocabulary Object Detection
This addresses the challenge of accurate and generalizable object detection for unseen categories in computer vision, representing a strong specific gain rather than an incremental improvement.
The paper tackles the problem of deriving reliable region-word alignment from image-text pairs for open-vocabulary object detection by proposing CoDet, which reformulates alignment as co-occurring object discovery, resulting in superior performance with 37.0 AP^m_novel and 44.7 AP^m_all on OV-LVIS, surpassing previous state-of-the-art by 4.2 and 9.8 points respectively.
Deriving reliable region-word alignment from image-text pairs is critical to learn object-level vision-language representations for open-vocabulary object detection. Existing methods typically rely on pre-trained or self-trained vision-language models for alignment, which are prone to limitations in localization accuracy or generalization capabilities. In this paper, we propose CoDet, a novel approach that overcomes the reliance on pre-aligned vision-language space by reformulating region-word alignment as a co-occurring object discovery problem. Intuitively, by grouping images that mention a shared concept in their captions, objects corresponding to the shared concept shall exhibit high co-occurrence among the group. CoDet then leverages visual similarities to discover the co-occurring objects and align them with the shared concept. Extensive experiments demonstrate that CoDet has superior performances and compelling scalability in open-vocabulary detection, e.g., by scaling up the visual backbone, CoDet achieves 37.0 $\text{AP}^m_{novel}$ and 44.7 $\text{AP}^m_{all}$ on OV-LVIS, surpassing the previous SoTA by 4.2 $\text{AP}^m_{novel}$ and 9.8 $\text{AP}^m_{all}$. Code is available at https://github.com/CVMI-Lab/CoDet.