A Black-Box Watermarking Modulation for Object Detection Models - Confiance.ai
Communication Dans Un Congrès Année : 2024

A Black-Box Watermarking Modulation for Object Detection Models

Mohammed Lansari
Lucas Mattioli
  • Fonction : Auteur
Boussad Addad
  • Fonction : Auteur
Paul-Marie Raffi
  • Fonction : Auteur
Martin Gonzalez
  • Fonction : Auteur
Mohamed Ibn Khedher
  • Fonction : Auteur

Résumé

Training a Deep Neural Network (DNN) from scratch comes with a substantial cost in terms of money, energy, data, and hardware. When such models are misused or redistributed without authorisation, the owner faces significant financial and intellectual property (IP) losses. Therefore, there is a pressing need to protect the IP of Machine Learning models to avoid these issues. ML watermarking emerges as a promising solution for model traceability. Watermarking has been well-studied for image classification models, but there is a significant research gap in its application to other tasks like object detection, for which no effective methods have been proposed yet. In this paper, we introduce a novel black-box watermarking method for object detection models. Our contributions include a watermarking technique that maps visual information to text semantics and a comparative study of fine-tuning techniques' impact on watermark detectability. We present the model's detection performance and evaluate fine-tuning strategies' effectiveness in preserving watermark integrity.
Fichier principal
Vignette du fichier
A_Black_Box_Watermarking_Modulation_for_Object_Detection_Models-2.pdf (937.77 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04726701 , version 1 (08-10-2024)

Identifiants

  • HAL Id : hal-04726701 , version 1

Citer

Mohammed Lansari, Lucas Mattioli, Boussad Addad, Paul-Marie Raffi, Katarzyna Kapusta, et al.. A Black-Box Watermarking Modulation for Object Detection Models. AI Trustworthiness and Risk Assessment for Challenged Contexts workshop (ATRACC). AAAI 2024 Fall Symposium, Nov 2024, Arlington, United States. ⟨hal-04726701⟩
0 Consultations
0 Téléchargements

Partager

More