Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Advancing High-Resolution Video-Language Representation with Large-Scale Video Transcriptions

About

We study joint video and language (VL) pre-training to enable cross-modality learning and benefit plentiful downstream VL tasks. Existing works either extract low-quality video features or learn limited text embedding, while neglecting that high-resolution videos and diversified semantics can significantly improve cross-modality learning. In this paper, we propose a novel High-resolution and Diversified VIdeo-LAnguage pre-training model (HD-VILA) for many visual tasks. In particular, we collect a large dataset with two distinct properties: 1) the first high-resolution dataset including 371.5k hours of 720p videos, and 2) the most diversified dataset covering 15 popular YouTube categories. To enable VL pre-training, we jointly optimize the HD-VILA model by a hybrid Transformer that learns rich spatiotemporal features, and a multimodal Transformer that enforces interactions of the learned video features with diversified texts. Our pre-training model achieves new state-of-the-art results in 10 VL understanding tasks and 2 more novel text-to-visual generation tasks. For example, we outperform SOTA models with relative increases of 40.4% R@1 in zero-shot MSR-VTT text-to-video retrieval task and 55.4% in high-resolution dataset LSMDC. The learned VL embedding is also effective in generating visually pleasing and semantically relevant results in text-to-visual editing and super-resolution tasks.

Hongwei Xue, Tiankai Hang, Yanhong Zeng, Yuchong Sun, Bei Liu, Huan Yang, Jianlong Fu, Baining Guo• 2021

Related benchmarks

TaskDatasetResultRank
Video Question AnsweringMSRVTT-QA
Accuracy40
481
Text-to-Video RetrievalDiDeMo (test)
R@128.8
376
Video Question AnsweringMSRVTT-QA (test)
Accuracy40
371
Text-to-Video RetrievalDiDeMo
R@10.288
360
Text-to-Video RetrievalMSR-VTT (test)
R@135.6
234
Text-to-Video RetrievalLSMDC (test)
R@117.4
225
Text-to-Video RetrievalMSR-VTT (1k-A)
R@1078
211
Text-to-Video RetrievalMSRVTT (test)
Recall@10.356
155
Video Question AnsweringTGIF-QA (test)--
89
Text-to-Video RetrievalActivityNet-captions (val1)
R@128.5
58
Showing 10 of 15 rows

Other info

Code

Follow for update