Prompt-guided DETR with RoI-pruned masked attention for open-vocabulary object detection

Prompt-OVD is an efficient and effective DETR-based framework for open-vocabulary object detection that utilizes class embeddings from CLIP as prompts, guiding the Transformer decoder to detect objects in base and novel classes. Additionally, our RoI-pruned masked attention helps leverage the zero-shot classification ability of the Vision Transformer-based CLIP, resulting in improved detection performance at a minimal computational cost. Our experiments on the OV-COCO and OV-LVIS datasets demonstrate that Prompt-OVD achieves an impressive 21.2 times faster inference speed than the first end-to-end open-vocabulary detection method (OV-DETR), while also achieving higher APs than four two-stage methods operating within similar inference time ranges. We release the code at https://github.com/DISL-Lab/Prompt-OVD. •A prompt-guided decoding is proposed to keep a constant number of object queries.•The decoding reduces the computational overhead of the Transformer decoder.•RoI-pruned masked attention benefits from a pre-trained CLIP at a minimal cost.•We improve the efficiency and accuracy of the end-to-end DETR-based OVD method.