paper-with-me

홈 › Papers

Spatial-temporal Concept based Explanation of 3D ConvNets

2022-06-09 · CVPR 2023 1 · Ying Ji, Yu Wang, Kensaku MORI, Jien Kato

Recent studies have achieved outstanding success in explaining 2D image recognition ConvNets. On the other hand, due to the computation cost and complexity of video data, the explanation of 3D video recognition ConvNets is relatively less studied. In this paper, we present a 3D ACE (Automatic Concept-based Explanation) framework for interpreting 3D ConvNets. In our approach: (1) videos are represented using high-level supervoxels, which is straightforward for human to understand; and (2) the interpreting framework estimates a score for each voxel, which reflects its importance in the decision procedure. Experiments show that our method can discover spatial-temporal concepts of different importance-levels, and thus can explore the influence of the concepts on a target task, such as action classification, in-depth. The codes are publicly available.

📄 PDF Abstract BibTeX arXiv:2206.05275

Code (1)

orangeeji/3d-ace 공식 구현 tf

Tasks

Action ClassificationVideo Recognition

Similar Papers 제목 키워드 기반

Initialization Strategies of Spatio-Temporal Convolutional Neural Networks

2015-03-25 · Elman Mansimov, Nitish Srivastava, Ruslan Salakhutdinov

We propose a new way of incorporating temporal information present in videos into Spatial Convolutional Neural Networks (ConvNets) trained on images, that avoids training Spatio-Temporal ConvNets from scratch. We describ…

Learning Spatiotemporal Features with 3D Convolutional Networks

2014-12-02 · ICCV 2015 12 · Du Tran, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani 외

We propose a simple, yet effective approach for spatiotemporal feature learning using deep 3-dimensional convolutional networks (3D ConvNets) trained on a large scale supervised video dataset. Our findings are three-fold…

Action RecognitionAction Recognition In VideosDynamic Facial Expression Recognition

Text Understanding from Scratch

2015-02-05 · Xiang Zhang, Yann Lecun

This article demontrates that we can apply deep learning to text understanding from character-level inputs all the way up to abstract text concepts, using temporal convolutional networks (ConvNets). We apply ConvNets to …

General ClassificationSentiment Analysis

Pooling the Convolutional Layers in Deep ConvNets for Action Recognition

2015-11-06 · Shichao Zhao, Yanbin Liu, Yahong Han, Richang Hong

Deep ConvNets have shown its good performance in image classification tasks. However it still remains as a problem in deep video representation for action recognition. The problem comes from two aspects: on one hand, cur…

Action Recognitionimage-classificationImage ClassificationTemporal Action Localization

Language-Bridged Spatial-Temporal Interaction for Referring Video Object Segmentation

2022-06-08 · CVPR 2022 1 · Zihan Ding, Tianrui Hui, Junshi Huang, Xiaoming Wei 외

Referring video object segmentation aims to predict foreground labels for objects referred by natural language expressions in videos. Previous methods either depend on 3D ConvNets or incorporate additional 2D ConvNets as…

DenoisingReferring Video Object SegmentationSemantic SegmentationVideo Object Segmentation+1