paper-with-me

Papers

ASCNet: Self-supervised Video Representation Learning with Appearance-Speed Consistency

2021-06-04 · ICCV 2021 10 · Deng Huang, Wenhao Wu, Weiwen Hu, Xu Liu, Dongliang He, Zhihua Wu, Xiangmiao Wu, Mingkui Tan, Errui Ding

We study self-supervised video representation learning, which is a challenging task due to 1) lack of labels for explicit supervision; 2) unstructured and noisy visual information. Existing methods mainly use contrastive loss with video clips as the instances and learn visual representation by discriminating instances from each other, but they need a careful treatment of negative pairs by either relying on large batch sizes, memory banks, extra modalities or customized mining strategies, which inevitably includes noisy data. In this paper, we observe that the consistency between positive samples is the key to learn robust video representation. Specifically, we propose two tasks to learn the appearance and speed consistency, respectively. The appearance consistency task aims to maximize the similarity between two clips of the same video with different playback speeds. The speed consistency task aims to maximize the similarity between two clips with the same playback speed but different appearance information. We show that optimizing the two tasks jointly consistently improves the performance on downstream tasks, e.g., action recognition and video retrieval. Remarkably, for action recognition on the UCF-101 dataset, we achieve 90.8\% accuracy without using any extra modalities or negative pairs for unsupervised pretraining, which outperforms the ImageNet supervised pretrained model. Codes and models will be available.

📄 PDF Abstract BibTeX arXiv:2106.02342

Code (0)

등록된 구현이 없습니다.

Tasks

Action RecognitionRepresentation LearningRetrievalVideo Retrieval

Similar Papers 제목 키워드 기반

Self-Supervised Video GANs: Learning for Appearance Consistency and Motion Coherency

2021-06-19 · CVPR 2021 1 · Sangeek Hyun, JiHwan Kim, Jae-Pil Heo

A video can be represented by the composition of appearance and motion. Appearance (or content) expresses the information invariant throughout time, and motion describes the time-variant movement. Here, we propose se…

Contrastive Learning

ASCNet: Asymmetric Sampling Correction Network for Infrared Image Destriping

2024-01-28 · Shuai Yuan, Hanlin Qin, Xiang Yan, Shiqi Yang 외

In a real-world infrared imaging system, effectively learning a consistent stripe noise removal model is essential. Most existing destriping methods cannot precisely reconstruct images due to cross-level semantic gaps an…

Feature UpsamplingImage Reconstruction

Multiview Pseudo-Labeling for Semi-supervised Learning from Video

2021-04-01 · ICCV 2021 10 · Bo Xiong, Haoqi Fan, Kristen Grauman, Christoph Feichtenhofer

We present a multiview pseudo-labeling approach to video learning, a novel framework that uses complementary views in the form of appearance and motion information for semi-supervised learning in video. The complementary…

Representation LearningVideo Recognition

Contrast and Order Representations for Video Self-Supervised Learning

2021-01-01 · ICCV 2021 10 · Kai Hu, Jie Shao, YuAn Liu, Bhiksha Raj 외

This paper studies the problem of learning self-supervised representations on videos. In contrast to image modality that only requires appearance information on objects or scenes, video needs to further explore the r…

Action RecognitionSelf-Supervised Action Recognition LinearSelf-Supervised Learning

Self-supervised Video Object Segmentation

2020-06-22 · Fangrui Zhu, Li Zhang, Yanwei Fu, Guodong Guo 외

The objective of this paper is self-supervised representation learning, with the goal of solving semi-supervised video object segmentation (a.k.a. dense tracking). We make the following contributions: (i) we propose to i…

ObjectOne-shot visual object segmentationRepresentation LearningSegmentation+4