Fully Convolutional Network and Region Proposal for Instance Identification with Egocentric Vision - Archive ouverte HAL Access content directly
Conference Papers Year :

Fully Convolutional Network and Region Proposal for Instance Identification with Egocentric Vision

Abstract

This paper presents a novel approach for egocentric image retrieval and object detection. This approach uses fully convolutional networks (FCN) to obtain region proposals without the need for an additional component in the network and training. It is particularly suited for small datasets with low object variability. The proposed network can be trained end-to-end and produces an effective global de-scriptor as an image representation. Additionally, it can be built upon any type of CNN pre-trained for classification. Through multiple experiments on two egocentric image datasets taken from museum visits, we show that the de-scriptor obtained using our proposed network outperforms those from previous state-of-the-art approaches. It is also just as memory-efficient, making it adapted to mobile devices such as an augmented museum audio-guide.
Fichier principal
Vignette du fichier
main.pdf (1.22 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01887959 , version 1 (04-10-2018)

Identifiers

Cite

Maxime Portaz, Matthias Kohl, Georges Quénot, Jean-Pierre Chevallet. Fully Convolutional Network and Region Proposal for Instance Identification with Egocentric Vision. IEEE International Conference on Computer Vision Workshop (ICCVW), Oct 2017, Venice, Italy. ⟨10.1109/ICCVW.2017.281⟩. ⟨hal-01887959⟩
31 View
61 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More