REVECA -- Rich Encoder-decoder framework for Video Event CAptioner
This work addresses the problem of generating descriptive captions for event boundaries in videos, which is important for applications in video understanding and analysis, though it appears incremental as it builds on existing methods like temporal segment networks and LoRA fine-tuning.
The paper tackled video event captioning by proposing REVECA, a framework that integrates spatial and temporal information to generate captions for event boundaries, achieving an average score of 50.97 on Kinetics-GEBC test data, which is a 10.17 improvement over the baseline.
We describe an approach used in the Generic Boundary Event Captioning challenge at the Long-Form Video Understanding Workshop held at CVPR 2022. We designed a Rich Encoder-decoder framework for Video Event CAptioner (REVECA) that utilizes spatial and temporal information from the video to generate a caption for the corresponding the event boundary. REVECA uses frame position embedding to incorporate information before and after the event boundary. Furthermore, it employs features extracted using the temporal segment network and temporal-based pairwise difference method to learn temporal information. A semantic segmentation mask for the attentional pooling process is adopted to learn the subject of an event. Finally, LoRA is applied to fine-tune the image encoder to enhance the learning efficiency. REVECA yielded an average score of 50.97 on the Kinetics-GEBC test data, which is an improvement of 10.17 over the baseline method. Our code is available in https://github.com/TooTouch/REVECA.