Communication Dans Un Congrès Année : 2025

Optimizing Frame Selection for Improved Video Quality Assessment Through Embedding Similarity

Résumé

This paper proposes a novel frame selection technique based on embedding similarity to optimize video quality assessment (VQA). By leveraging high-dimensional feature embeddings extracted from deep neural networks (ResNet-50, VGG-16, and CLIP), we introduce a similarity-preserving approach that prioritizes perceptually relevant frames while reducing redundancy. The proposed method is evaluated on two datasets, CVD2014 and KonViD-1k, demonstrating robust performance across synthetic and real-world distortions. Results show that the proposed approach outperforms state-of-the-art methods, particularly in handling diverse and in-the-wild video content, achieving robust performances on KonViD-1k. This work highlights the importance of embedding-driven frame selection in improving the accuracy and efficiency of VQA methods.

Fichier principal
Vignette du fichier
EI_2025_IQSP-251_Abderrezzaq--Sendjasni.pdf (11.04 Mo) Télécharger le fichier
Origine Fichiers éditeurs autorisés sur une archive ouverte
Licence

Dates et versions

hal-05120265 , version 1 (19-06-2025)

Licence

Identifiants

Citer

Abderrezzaq Sendjasni, Mohamed-Chaker Larabi, Seif-Eddine Benkabou. Optimizing Frame Selection for Improved Video Quality Assessment Through Embedding Similarity. 2025 Electronic Imaging Symposium : Image Quality and System Performance, Society for Imaging Science and Technology (IS&T), Feb 2025, Burlingame (California), United States. pp.251, ⟨10.2352/ei.2025.37.9.iqsp-251⟩. ⟨hal-05120265⟩
376 Consultations
145 Téléchargements

Altmetric

Partager

  • More