VisTa: Visual-contextual and Text-augmented Zero-shot Object-level OOD Detection.

Bin Zhang, Xiaoyang Qu, Guokuan Li, Jiguang Wan, Jianzong Wang

Browse the full ICASSP paper archive.