HEVC stream fingerprinting for organic video: proof of concepts and illustrations

Mohamed Allouche; Mihai Mitrea; Carl De Sousa Trias

doi:10.2352/EI.2025.37.10.IPAS-233

HEVC stream fingerprinting for organic video: proof of concepts and illustrations

Mohamed Allouche
, Mihai Mitrea
, Carl De Sousa Trias

Telecom Sudparis

Research output: Contribution to journal › Conference article › peer-review

Abstract

While conventional video fingerprinting methods act in the uncompressed domain (pixels and/or directly derived representations from pixels), the present paper establishes the proof of concepts for compressed domain video fingerprinting. Thus, visual content is processed at the level of compressed stream syntax elements (luma/chroma coefficients, and intraprediction modes) by a homemade NN-based solution backboned by conventional CNN models (ResNet and MobileNet). The experimental validations are obtained out of processing a state of the art and a homemade HEVC compressed video databases, and bring forth Accuracy, Precision and Recall values larger than 0.9.

Original language	English
Article number	IPAS-233
Journal	IS and T International Symposium on Electronic Imaging Science and Technology
Volume	37
Issue number	10
DOIs	https://doi.org/10.2352/EI.2025.37.10.IPAS-233
Publication status	Published - 1 Jan 2025
Event	IS and T International Symposium on Electronic Imaging 2025: 23rd Image Processing: Algorithms and Systems, IPAS 2025 - Burlingame, United States Duration: 2 Feb 2025 → 6 Feb 2025

Access to Document

10.2352/EI.2025.37.10.IPAS-233

Cite this

@article{37d4e904e5a2458b8ad5af3dd4d2c531,

title = "HEVC stream fingerprinting for organic video: proof of concepts and illustrations",

abstract = "While conventional video fingerprinting methods act in the uncompressed domain (pixels and/or directly derived representations from pixels), the present paper establishes the proof of concepts for compressed domain video fingerprinting. Thus, visual content is processed at the level of compressed stream syntax elements (luma/chroma coefficients, and intraprediction modes) by a homemade NN-based solution backboned by conventional CNN models (ResNet and MobileNet). The experimental validations are obtained out of processing a state of the art and a homemade HEVC compressed video databases, and bring forth Accuracy, Precision and Recall values larger than 0.9.",

author = "Mohamed Allouche and Mihai Mitrea and Trias, \{Carl De Sousa\}",

note = "Publisher Copyright: {\textcopyright} 2025 Society for Imaging Science and Technology.; IS and T International Symposium on Electronic Imaging 2025: 23rd Image Processing: Algorithms and Systems, IPAS 2025 ; Conference date: 02-02-2025 Through 06-02-2025",

year = "2025",

month = jan,

day = "1",

doi = "10.2352/EI.2025.37.10.IPAS-233",

language = "English",

volume = "37",

journal = "IS and T International Symposium on Electronic Imaging Science and Technology",

issn = "2470-1173",

number = "10",

}

TY - JOUR

T1 - HEVC stream fingerprinting for organic video

T2 - IS and T International Symposium on Electronic Imaging 2025: 23rd Image Processing: Algorithms and Systems, IPAS 2025

AU - Allouche, Mohamed

AU - Mitrea, Mihai

AU - Trias, Carl De Sousa

PY - 2025/1/1

Y1 - 2025/1/1

N2 - While conventional video fingerprinting methods act in the uncompressed domain (pixels and/or directly derived representations from pixels), the present paper establishes the proof of concepts for compressed domain video fingerprinting. Thus, visual content is processed at the level of compressed stream syntax elements (luma/chroma coefficients, and intraprediction modes) by a homemade NN-based solution backboned by conventional CNN models (ResNet and MobileNet). The experimental validations are obtained out of processing a state of the art and a homemade HEVC compressed video databases, and bring forth Accuracy, Precision and Recall values larger than 0.9.

AB - While conventional video fingerprinting methods act in the uncompressed domain (pixels and/or directly derived representations from pixels), the present paper establishes the proof of concepts for compressed domain video fingerprinting. Thus, visual content is processed at the level of compressed stream syntax elements (luma/chroma coefficients, and intraprediction modes) by a homemade NN-based solution backboned by conventional CNN models (ResNet and MobileNet). The experimental validations are obtained out of processing a state of the art and a homemade HEVC compressed video databases, and bring forth Accuracy, Precision and Recall values larger than 0.9.

UR - https://www.scopus.com/pages/publications/105000822656

U2 - 10.2352/EI.2025.37.10.IPAS-233

DO - 10.2352/EI.2025.37.10.IPAS-233

M3 - Conference article

AN - SCOPUS:105000822656

SN - 2470-1173

VL - 37

JO - IS and T International Symposium on Electronic Imaging Science and Technology

JF - IS and T International Symposium on Electronic Imaging Science and Technology

IS - 10

M1 - IPAS-233

Y2 - 2 February 2025 through 6 February 2025

ER -

HEVC stream fingerprinting for organic video: proof of concepts and illustrations

Abstract

Access to Document

Other files and links

Fingerprint

Cite this