☆ 4.5 Article

NOVAction23: Addressing the data diversity gap by uniquely generated synthetic sequences for real-world human action recognition

COMPUTERS & GRAPHICS-UK (2024)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Engineering, Electrical & Electronic

COCAS plus : Large-Scale Clothes-Changing Person Re-Identification With Clothes Templates

Shihua Li et al.

Summary: In recent years, person re-identification (ReID) has seen rapid development due to its practical applications. However, most existing benchmarks assume that the same person wears the same clothes in captured images, which is not realistic. To address this issue, the Clothes-Changing person ReID (CC-ReID) problem is introduced. In this paper, a new task called Clothes Template based Clothes-Changing person ReID (CTCC-ReID) is proposed, and a large-scale dataset named ClOthes ChAnging person Set Plus (COCAS+) is constructed. Furthermore, a Dual-Attention Biometric-Clothes Transfusion Network (DualBCT-Net) is proposed for CTCC-ReID, showing significant improvements over state-of-the-art methods.

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY (2023)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Revisiting Skeleton-based Action Recognition

Haodong Duan et al.

Summary: This paper proposes a new approach called PoseConv3D for skeleton-based action recognition. Compared to GCN-based methods, PoseConv3D is more effective in learning spatiotemporal features, more robust against pose estimation noises, and generalizes better in cross-dataset settings. It can also handle multiple-person scenarios without additional computation costs and can be easily integrated with other modalities.

2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022) (2022)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Arbitrary-view human action recognition via novel-view action generation

Kumie Gedamu et al.

Summary: In this study, a Two Branch Novel-View action Generation approach based on auxiliary conditional GAN is proposed to generate novel-view action samples for enlarging the view range in the training set. Furthermore, a view-domain generalization model is introduced to improve the recognition performance of arbitrary-view human action recognition by narrowing the representation of actions in different views. The proposed approach achieves outstanding performance in human action recognition on three large-scale RGB+D skeleton datasets.

PATTERN RECOGNITION (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Synthetic Humans for Action Recognition from Unseen Viewpoints

Gul Varol et al.

Summary: This research explores the potential of synthetic training data for RGB human action recognition, significantly improving performance through investigations on variations and augmentations.

INTERNATIONAL JOURNAL OF COMPUTER VISION (2021)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Using synthetic data for person tracking under adverse weather conditions

Abdulrahman Kerim et al.

Summary: Robust visual tracking is crucial in various fields, but still needs improvement under adverse weather conditions. Researchers introduced new real-world and synthetic person tracking datasets, showing that incorporating synthetic data into training sequences can enhance the performance of state-of-the-art deep trackers in adverse weather conditions.

IMAGE AND VISION COMPUTING (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Synthetic18K: Learning better representations for person re-ID and attribute recognition from 1.4 million synthetic images

Onur Can Uner et al.

Summary: The paper proposes using synthetic person images to learn robust representations, achieving significant improvements in prediction performances for person re-identification and attribute recognition tasks. Pretraining simple deep architectures on Synthetic18K and fine-tuning on real data leads to better prediction results than or comparable to state-of-the-art models.

SIGNAL PROCESSING-IMAGE COMMUNICATION (2021)

添加到收藏夹

Article Engineering, Civil

Enhancing Data-Driven Algorithms for Human Pose Estimation and Action Recognition Through Simulation

Dennis Ludl et al.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2020)

添加到收藏夹

Article Computer Science, Artificial Intelligence

NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understanding

Jun Liu et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2020)

添加到收藏夹

Article Computer Science, Theory & Methods

A survey on Image Data Augmentation for Deep Learning

Connor Shorten et al.

JOURNAL OF BIG DATA (2019)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence

Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset

Joao Carreira et al.

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017) (2017)

添加到收藏夹

Article Computer Science, Artificial Intelligence

A novel 2D/3D database with automatic face annotation for head tracking and pose estimation

Mikel Ariz et al.

COMPUTER VISION AND IMAGE UNDERSTANDING (2016)

添加到收藏夹

Article Computer Science, Artificial Intelligence

Histogram of Oriented Principal Components for Cross-View Action Recognition

Hossein Rahmani et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2016)

添加到收藏夹

Proceedings Paper Computer Science, Information Systems

A Multi-modal & Multi-view & Interactive Benchmark Dataset for Human Action Recognition

Ning Xu et al.

MM'15: PROCEEDINGS OF THE 2015 ACM MULTIMEDIA CONFERENCE (2015)

添加到收藏夹

Article Computer Science, Artificial Intelligence

3D Convolutional Neural Networks for Human Action Recognition

Shuiwang Ji et al.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2013)

添加到收藏夹

Article Education & Educational Research

Conducting Video Research in the Learning Sciences: Guidance on Selection, Analysis, Technology, and Ethics

Sharon J. Derry et al.

JOURNAL OF THE LEARNING SCIENCES (2010)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.