[HTML][HTML] Survey on videos data augmentation for deep learning models

N Cauli, D Reforgiato Recupero - Future Internet, 2022‏ - mdpi.com
In most Computer Vision applications, Deep Learning models achieve state-of-the-art
performances. One drawback of Deep Learning is the large amount of data needed to train …

Towards discriminative representation learning for unsupervised person re-identification

T Isobe, D Li, L Tian, W Chen… - Proceedings of the …, 2021‏ - openaccess.thecvf.com
In this work, we address the problem of unsupervised domain adaptation for person re-ID
where annotations are available for the source domain but not for target. Previous methods …

Compression-aware video super-resolution

Y Wang, T Isobe, X Jia, X Tao… - Proceedings of the …, 2023‏ - openaccess.thecvf.com
Videos stored on mobile devices or delivered on the Internet are usually in compressed
format and are of various unknown compression parameters, but most video super …

Delving into probabilistic uncertainty for unsupervised domain adaptive person re-identification

J Han, YL Li, S Wang - Proceedings of the AAAI conference on artificial …, 2022‏ - ojs.aaai.org
Clustering-based unsupervised domain adaptive (UDA) person re-identification (ReID)
reduces exhaustive annotations. However, owing to unsatisfactory feature embedding and …

Benchmarking the robustness of temporal action detection models against temporal corruptions

R Zeng, X Chen, J Liang, H Wu… - Proceedings of the …, 2024‏ - openaccess.thecvf.com
Temporal action detection (TAD) aims to locate action positions and recognize action
categories in long-term untrimmed videos. Although many methods have achieved …

A feature-space multimodal data augmentation technique for text-video retrieval

A Falcon, G Serra, O Lanz - Proceedings of the 30th ACM international …, 2022‏ - dl.acm.org
Every hour, huge amounts of visual contents are posted on social media and user-
generated content platforms. To find relevant videos by means of a natural language query …

Markov game video augmentation for action segmentation

N Aziere, S Todorovic - Proceedings of the IEEE/CVF …, 2023‏ - openaccess.thecvf.com
This paper addresses data augmentation for action segmentation. Our key novelty is that we
augment the original training videos in the deep feature space, not in the visual …

Shapeaug: Occlusion augmentation for event camera data

K Bendig, R Schuster, D Stricker - arxiv preprint arxiv:2401.02274, 2024‏ - arxiv.org
Recently, Dynamic Vision Sensors (DVSs) sparked a lot of interest due to their inherent
advantages over conventional RGB cameras. These advantages include a low latency, a …

Enhanced Heart Disease Classification Using Dual Attention Mechanisms and 3D-Echo Fusion Algorithm in Echocardiogram Videos

S Deepika, N Jaisankar - IEEE Access, 2024‏ - ieeexplore.ieee.org
Heart disease remains a leading cause of mortality worldwide, making early detection and
diagnosis crucial for preventing severe outcomes. Echocardiogram based classification of …

Group RandAugment: Video augmentation for action recognition

F An, B Zhang, Z Wang, W Dong… - 2022 5th International …, 2022‏ - ieeexplore.ieee.org
Data augmentation, as a critical strategy in deep learning, well improves the sample
diversity for network training, leading to the obvious improvement of model generalization …