Deriving reliable region-word alignment from image-text pairs is critical to learnobject-level vision-language representations for open-vocabulary object detection.Existing methods typically rely on pre-trained or self-trained vision-languagemodels for alignment, which are prone to limitations in localization accuracy orgeneralization capabilities. In this paper, we propose CoDet, a novel approachthat overcomes the reliance on pre-aligned vision-language space by reformulatingregion-word alignment as a co-occurring object discovery problem. Intuitively, bygrouping 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 andalign them with the shared concept. Extensive experiments demonstrate that CoDethas superior performances and compelling scalability in open-vocabulary detection,e.g., by scaling up the visual backbone, CoDet achieves 37.0 $AP^m_{novel}$ and 44.7 $AP^m_{all}$ on OV-LVIS, surpassing the previous SoTA by 4.2 $AP^m_{novel}$ and 9.8 $AP^m_{all}$. Code is available at https://github.com/CVMI-Lab/CoDet.

Recent studies on generalizable object detection have attracted increasing attention with additional weak supervision from large-scale datasets with image-level labels.However, weakly-supervised detection learning often suffers from image-to-box label mismatch, i.e., image-levellabels do not convey precise object information.We design Language Hierarchical Self-training (LHST) that introduces language hierarchy into weakly-supervised detector training for learning more generalizable detectors.LHST expands the image-level labels with language hierarchy and enables co-regularization between the expanded labels and self-training. Specifically, the expanded labels regularize self-training by providing richer supervision and mitigating the image-to-box label mismatch, while self-training allows assessing and selecting the expanded labels according to the predicted reliability.In addition, we design language hierarchical prompt generation that introduces language hierarchy into prompt generation which helps bridge the vocabulary gaps between training and testing.Extensive experiments show that the proposed techniques achieve superior generalization performance consistently across 14 widely studied object detection datasets.

Deriving reliable region-word alignment from image-text pairs is critical to learnobject-level vision-language representations for open-vocabulary object detection.Existing methods typically rely on pre-trained or self-trained vision-languagemodels for alignment, which are prone to limitations in localization accuracy orgeneralization capabilities. In this paper, we propose CoDet, a novel approachthat overcomes the reliance on pre-aligned vision-language space by reformulatingregion-word alignment as a co-occurring object discovery problem. Intuitively, bygrouping 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 andalign them with the shared concept. Code is available at https://github.com/CVMI-Lab/CoDet.

Recent open-vocabulary detection methods aim to detect novel objects by distilling knowledge from vision-language models (VLMs) trained on a vast amount of image-text pairs. To improve the effectiveness of these methods, researchers have utilized datasets with a large vocabulary that contains a large number of object classes, under the assumption that such data will enable models to extract comprehensive knowledge on the relationships between various objects and better generalize to unseen object classes. In this study, we argue that more fine-grained labels are necessary to extract richer knowledge about novel objects, including object attributes and relationships, in addition to their names. To address this challenge, we propose a simple and effective method named Pseudo Caption Labeling (PCL), which utilizes an image captioning model to generate captions that describe object instances from diverse perspectives. The resulting pseudo caption labels offer dense samples for knowledge distillation. On the LVIS benchmark, our best model trained on the de-duplicated VisualGenome dataset achieves an AP of 34.5 and an APr of 30.6, comparable to the state-of-the-art performance. PCL's simplicity and flexibility are other notable features, as it is a straightforward pre-processing technique that can be used with any image captioning model without imposing any restrictions on model architecture or training process.