Unifying Event Detection and Captioning as Sequence Generation via Pre-training.

Qi Zhang, Yuqing Song, Qin Jin

Browse the full ECCV paper archive.