Salvato in:
Dettagli Bibliografici
Autori principali: Sun, Jianqiao, Su, Yudi, Zhang, Hao, Cheng, Ziheng, Zeng, Zequn, Wang, Zhengjue, Chen, Bo, Yuan, Xin
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2401.04903
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909067527585792
author Sun, Jianqiao
Su, Yudi
Zhang, Hao
Cheng, Ziheng
Zeng, Zequn
Wang, Zhengjue
Chen, Bo
Yuan, Xin
author_facet Sun, Jianqiao
Su, Yudi
Zhang, Hao
Cheng, Ziheng
Zeng, Zequn
Wang, Zhengjue
Chen, Bo
Yuan, Xin
contents Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for storage and transmission. However, in such a pipeline, some potential shortcomings are inevitable, i.e., information redundancy resulting in low efficiency and information loss during the sampling process for captioning. To address these problems, in this paper, we propose a novel VC pipeline to generate captions directly from the compressed measurement, which can be captured by a snapshot compressive sensing camera and we dub our model SnapCap. To be more specific, benefiting from the signal simulation, we have access to obtain abundant measurement-video-annotation data pairs for our model. Besides, to better extract language-related visual representations from the compressed measurement, we propose to distill the knowledge from videos via a pre-trained CLIP with plentiful language-vision associations to guide the learning of our SnapCap. To demonstrate the effectiveness of SnapCap, we conduct experiments on two widely-used VC datasets. Both the qualitative and quantitative results verify the superiority of our pipeline over conventional VC pipelines. In particular, compared to the "caption-after-reconstruction" methods, our SnapCap can run at least 3$\times$ faster, and achieve better caption results.
format Preprint
id arxiv_https___arxiv_org_abs_2401_04903
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SnapCap: Efficient Snapshot Compressive Video Captioning
Sun, Jianqiao
Su, Yudi
Zhang, Hao
Cheng, Ziheng
Zeng, Zequn
Wang, Zhengjue
Chen, Bo
Yuan, Xin
Computer Vision and Pattern Recognition
Video Captioning (VC) is a challenging multi-modal task since it requires describing the scene in language by understanding various and complex videos. For machines, the traditional VC follows the "imaging-compression-decoding-and-then-captioning" pipeline, where compression is pivot for storage and transmission. However, in such a pipeline, some potential shortcomings are inevitable, i.e., information redundancy resulting in low efficiency and information loss during the sampling process for captioning. To address these problems, in this paper, we propose a novel VC pipeline to generate captions directly from the compressed measurement, which can be captured by a snapshot compressive sensing camera and we dub our model SnapCap. To be more specific, benefiting from the signal simulation, we have access to obtain abundant measurement-video-annotation data pairs for our model. Besides, to better extract language-related visual representations from the compressed measurement, we propose to distill the knowledge from videos via a pre-trained CLIP with plentiful language-vision associations to guide the learning of our SnapCap. To demonstrate the effectiveness of SnapCap, we conduct experiments on two widely-used VC datasets. Both the qualitative and quantitative results verify the superiority of our pipeline over conventional VC pipelines. In particular, compared to the "caption-after-reconstruction" methods, our SnapCap can run at least 3$\times$ faster, and achieve better caption results.
title SnapCap: Efficient Snapshot Compressive Video Captioning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2401.04903