Deep frame interpolation for video compression

Abstract : Deep neural networks have been recently proposed to solve video interpolation tasks. Given a past and future frame, such networks can be trained to successfully predict the intermediate frame(s). In the context of video compression, these architectures could be useful as an additional inter-prediction mode. Current inter-prediction methods rely on block-matching techniques to estimate the motion between consecutive frames. This approach has severe limitations for handling complex non-translational motions, and is still limited to block-based motion vectors. This paper presents a deep frame interpolation network for video compression aiming at solving the previous limitations, i.e. able to cope with all types of geometrical deformations by providing a dense motion compensation. Experiments with the classical bi-directional hierarchical video coding structure demonstrate the efficiency of the proposed approach over the traditional tools of the HEVC codec.
Document type :
Conference papers
Complete list of metadatas

Cited literature [26 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02202172
Contributor : Christine Guillemot <>
Submitted on : Wednesday, July 31, 2019 - 4:01:24 PM
Last modification on : Friday, August 2, 2019 - 2:24:45 AM

File

BegaintGGG19.pdf
Files produced by the author(s)

Identifiers

Citation

Jean Bégaint, Franck Galpin, Philippe Guillotel, Christine Guillemot. Deep frame interpolation for video compression. DCC 2019 - Data Compression Conference, Mar 2019, Snowbird, United States. ⟨10.1109/DCC.2019.00068⟩. ⟨hal-02202172⟩

Share

Metrics

Record views

50

Files downloads

16