Skip to Main content Skip to Navigation
Journal articles

CNN-based temporal detection of motion saliency in videos

Abstract : The problem addressed in this paper appertains to the domain of motion saliency in videos. However this is a new problem since we aim to extract the temporal segments of the video where motion saliency is present. It turns out to be a frame-based classification problem. A frame will be classified as dynamically salient if it contains local motion departing from its context. Temporal motion saliency detection is relevant for applications where one needs to trigger alerts or to monitor dynamic behaviours from videos. It can also be viewed as a prerequisite before computing motion saliency maps. The proposed approach handles situations with a mobile camera. It involves two main stages consisting first in cancelling the global motion due to the camera movement, then in applying a deep learning classification framework. We have investigated two ways of implementing the first stage, based on image warping, and on residual flow respectively. Experiments on real videos demonstrate that we can obtain accurate classification in highly challenging situations.
Document type :
Journal articles
Complete list of metadatas

Cited literature [40 references]  Display  Hide  Download

https://hal.inria.fr/hal-02345209
Contributor : Léo Maczyta <>
Submitted on : Monday, November 4, 2019 - 2:16:52 PM
Last modification on : Friday, July 10, 2020 - 4:24:49 PM
Long-term archiving on: : Wednesday, February 5, 2020 - 11:04:34 PM

File

CNN-based temporal detection o...
Files produced by the author(s)

Identifiers

Citation

Léo Maczyta, Patrick Bouthemy, Olivier Le Meur. CNN-based temporal detection of motion saliency in videos. Pattern Recognition Letters, Elsevier, 2019, 128, pp.298-305. ⟨10.1016/j.patrec.2019.09.016⟩. ⟨hal-02345209⟩

Share

Metrics

Record views

122

Files downloads

1289