Context-adaptive neural network based prediction for image compression

Thierry Dumas 1 Aline Roumy 1 Christine Guillemot 1
1 Sirocco - Analysis representation, compression and communication of visual data
Inria Rennes – Bretagne Atlantique , IRISA_D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : This paper describes a set of neural network architectures, called Prediction Neural Networks Set (PNNS), based on both fully-connected and convolutional neural networks, for intra image prediction. The choice of neural network for predicting a given image block depends on the block size, hence does not need to be signalled to the decoder. It is shown that, while fully-connected neural networks give good performance for small block sizes, convolutional neural networks provide better predictions in large blocks with complex textures. Thanks to the use of masks of random sizes during training, the neural networks of PNNS well adapt to the available context that may vary, depending on the position of the image block to be predicted. When integrating PNNS into a H.265 codec, PSNR-rate performance gains going from 1.46% to 5.20% are obtained. These gains are on average 0.99% larger than those of prior neural network based methods. Unlike the H.265 intra prediction modes, which are each specialized in predicting a specific texture, the proposed PNNS can model a large set of complex textures.
Complete list of metadatas

Cited literature [52 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01841034
Contributor : Thierry Dumas <>
Submitted on : Thursday, August 8, 2019 - 10:45:50 PM
Last modification on : Sunday, August 11, 2019 - 1:09:55 AM

File

context_adaptive_neural.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01841034, version 2
  • ARXIV : 1807.06244

Citation

Thierry Dumas, Aline Roumy, Christine Guillemot. Context-adaptive neural network based prediction for image compression. 2019. ⟨hal-01841034v2⟩

Share

Metrics

Record views

48

Files downloads

54