Communication Dans Un Congrès Année : 2017

Detecting Humans in RGB-D Data with CNNs

Résumé

We address the problem of people detection in RGB-D data where we leverage depth information to develop a region-of-interest (ROI) selection method that provides proposals to two color and depth CNNs. To combine the detections produced by the two CNNs, we propose a novel fusion approach based on the characteristics of depth images. We also present a new depth-encoding scheme, which not only encodes depth images into three channels but also enhances the information for classification. We conduct experiments on a publicly available RGB-D people dataset and show that our approach outperforms the baseline models that only use RGB data.

Fichier principal
Vignette du fichier
2207.08064.pdf (4.74 Mo) Télécharger le fichier

Dates et versions

hal-02549879 , version 1 (13-03-2024)

Licence

Identifiants

  • HAL Id : hal-02549879 , version 1

Citer

Kaiyang Zhou, Adeline Paiement, Majid Mirmehdi. Detecting Humans in RGB-D Data with CNNs. IAPR International Conference on Machine Vision Applications, 2017, Nagoya, Japan. ⟨hal-02549879⟩

Collections

28 Consultations
60 Téléchargements

Partager

  • More