51,706 Hits in 4.6 sec

Machine Learning based Post Processing Artifact Reduction in HEVC Intra Coding [article]

K.R. Rao, Ninad Gorey
2019 arXiv   pre-print
These operations consumes extra signaling bit and becomes an overhead to network. In this paper we proposed a new Deep learning based algorithm on SAO filtering operation.  ...  We also demonstrated that deeper architecture model can effectively be trained with the features learnt in a shallow network using data augmentation and transfer learning based techniques.  ...  Fig. 1 shows our decoder block diagram with newly added SDCNN network. Our CNN based network takes the compressed image as input from HEVC encoder and output reconstructed residual image.  ... 
arXiv:1912.13100v1 fatcat:rljhvmpfxze5bcreb54maokjsi

Image and Video Compression with Neural Networks: A Review

Siwei Ma, Xinfeng Zhang, Chuanmin Jia, Zhenghui Zhao, Shiqi Wang, Shanshe Wanga
2019 IEEE transactions on circuits and systems for video technology (Print)  
Moreover, the end-to-end image and video coding frameworks based on neural networks are also reviewed, revealing interesting explorations on next generation image and video coding frameworks/standards.  ...  The evolution and development of neural network based compression methodologies are introduced for images and video respectively.  ...  In [104] , they provided an efficient solution for CNN based loop filters with memory efficiency.  ... 
doi:10.1109/tcsvt.2019.2910119 fatcat:ibwmmewdlfcexjxfetsxzga52y

Learning a Virtual Codec Based on Deep Convolutional Neural Network to Compress Image [article]

Lijun Zhao, Huihui Bai, Anhong Wang, Yao Zhao
2018 arXiv   pre-print
In this paper, we propose an end-to-end image compression framework based on convolutional neural network to resolve the problem of non-differentiability of the quantization function in the standard codec  ...  Meanwhile, an advanced learning algorithm is proposed to train our deep neural networks for compression.  ...  Training details Our framework of learning a virtual codec neural network to compress image is implemented with TensorFlow [38] .  ... 
arXiv:1712.05969v7 fatcat:5so7dj3xzra4rkzcmwn3pdnb6m

Virtual Codec Supervised Re-Sampling Network for Image Compression [article]

Lijun Zhao, Huihui Bai, Anhong Wang, Yao Zhao
2018 arXiv   pre-print
We generalize this method for standard-compliant image compression (SCIC) framework and deep neural networks based compression (DNNC) framework.  ...  In this paper, we propose an image re-sampling compression method by learning virtual codec network (VCN) to resolve the non-differentiable problem of quantization function for image compression.  ...  In [35] , self-learning based image decomposition is applied for single image de-noising with an over-complete dictionary, which can be used to alleviate coding artifacts.  ... 
arXiv:1806.08514v2 fatcat:roos7bbw6rhexko6gnc4dvyfau

An End-to-End Compression Framework Based on Convolutional Neural Networks [article]

Feng Jiang, Wen Tao, Shaohui Liu, Jie Ren, Xun Guo, Debin Zhao
2017 arXiv   pre-print
Such a design also makes the proposed compression framework compatible with existing image coding standards.  ...  To achieve high-quality image compression at low bit rates, two CNNs are seamlessly integrated into an end-to-end compression framework.  ...  Image Compression Based on Deep Learning Recently, deep learning has been used both for lossy and lossless image compression and achieved competitive performance.  ... 
arXiv:1708.00838v1 fatcat:xr6yssotmrbxzm47lyebdhfvye

Guest Editorial Introduction to Special Section on Learning-Based Image and Video Compression

Shan Liu, Wen-Hsiao Peng, Lu Yu
2020 IEEE transactions on circuits and systems for video technology (Print)  
issued another CfE on Deep Learning-Based Image Compression with some different emphases and considerations.  ...  This Special Section consists of 12 articles with one emphasizing end-to-end learning-based image compression, and the rest focusing on learning-based coding tools.  ... 
doi:10.1109/tcsvt.2020.2995955 fatcat:qv3h5hpjq5gu7l324mbzwjjjmm

Multiple Description Coding Based on Convolutional Auto-encoder

Hongfei Li, Lili Meng, Jia Zhang, Yanyan Tan, Yuwei Ren, Huaxiang Zhang
2019 IEEE Access  
Deep learning, such as convolutional neural networks, has been achieved great success in image processing, computer vision task, and image compression, and has achieved better performance.  ...  This paper designs a multiple description coding frameworks based on symmetric convolutional auto-encoder, which can achieve high-quality image reconstruction.  ...  In [4] , an efficient end-to-end compression framework based on two CNNs is proposed, the first CNN is used to generate a compact intermediate representation for encoding using an image encoder, the second  ... 
doi:10.1109/access.2019.2900498 fatcat:5c6wlwecerfoxfoccpxvnbgbyq

A Unified Framework for Encryption and Decryption of Images Based on Autoencoder (UFED)

2021 International Journal of Advanced Trends in Computer Science and Engineering  
Our proposed unified framework for encryption and decryption of images based on an autoencoder (UFED) can control the cost during encryption and decryption using modern techniques like deep learning and  ...  We achieved the best image-compression ratio with Autoencoder over JPEG; JPEG typically achieves 10:1 compression with little perceptible loss in image quality.  ...  In November 2019, the article presented an architecture that works based on the D.C.T. and compressed the JPEG image with the help of VHDL [22] .  ... 
doi:10.30534/ijatcse/2021/451032021 fatcat:6ykv4j3bnnbgzbub7uop46r7gi

Preprocessing Enhanced Image Compression for Machine Vision [article]

Guo Lu, Xingtong Ge, Tianxiong Zhong, Jing Geng, Qiang Hu
2022 arXiv   pre-print
Instead of relying on the learned image codecs for end-to-end optimization, our framework is built upon the traditional non-differential codecs, which means it is standard compatible and can be easily  ...  We provide extensive experiments by evaluating our compression method for two representative downstream tasks with different backbone networks.  ...  Proxy Network In our framework, to enable an end-to-end optimization for the whole system, a learned image compression network is introduced as the proxy network to replace the traditional codec during  ... 
arXiv:2206.05650v1 fatcat:lnpdgzowpzgxlfvug2wzmion34

Deep Convolution Networks for Compression Artifacts Reduction [article]

Ke Yu, Chao Dong, Chen Change Loy, Xiaoou Tang
2016 arXiv   pre-print
This also leads to a more general CNN framework that has a close relationship with the conventional Multi-Layer Perceptron (MLP).  ...  We also demonstrate that a deeper model can be effectively trained with features learned in a shallow network.  ...  based RTF [13] , on restoring JPEG-compressed images.  ... 
arXiv:1608.02778v1 fatcat:okb4eqezabcgzif3yxbszwdq74

Deep Face Model Compression Using Entropy-Based Filter Selection [chapter]

Bingbing Han, Zhihong Zhang, Chuanyu Xu, Beizhan Wang, Guosheng Hu, Lu Bai, Qingqi Hong, Edwin R. Hancock
2017 Lecture Notes in Computer Science  
First the importance of each filter in each layer is evaluated by our entropy-based method. Then some unimportant filters are removed according to a predefined compressing rate.  ...  In this work, we propose an entropy-based prune metric to reduce the size of intermediate activations so as to accelerate and compress CNN models both in training and inference stages.  ...  Our framework can reduce the number of filters so as to compress the size of activation in each layer, which catch almost no attention in previous work. • We use an effective learning strategy to make  ... 
doi:10.1007/978-3-319-68548-9_12 fatcat:3pqddtwdsbbepkhnvjqjo62bqm

Reconstruction of Compressively Sensed Images using Convex Tikhonov Sparse Dictionary Learning and Adaptive Spectral Filtering [article]

Nishant Deepak Keni, Amol Mangirish Singbal, Rizwan Ahmed
2019 arXiv   pre-print
Once the image is reconstructed from the compressively sensed samples, we use adaptive frequency and spatial filtering techniques to move towards exact image recovery.  ...  Dictionary learning algorithms have made it possible to engineer data adaptive dictionaries which have promising applications in image compression and image enhancement.  ...  In this article, we have proposed an algorithm for image recovery of compressively sensed images using regularized convex sparse dictionary learning and adaptive spectral filtering.  ... 
arXiv:1801.09135v2 fatcat:cufxwsrievcztif26kwoyksv4e

A Hardware Prototype Targeting Distributed Deep Learning for On-device Inference

Alien-Jasmin Farcas, Guihong Li, Kartikeya Bhardwaj, Radu Marculescu
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
This is an important step towards deploying deep learning models for IoT applications.  ...  This paper presents a hardware prototype and a framework for a new communication-aware model compression for distributed on-device inference.  ...  Approach Our framework (Fig. 1 .) starts with an initial teacher model that is pre-trained.  ... 
doi:10.1109/cvprw50498.2020.00207 dblp:conf/cvpr/FarcasLBM20 fatcat:posjk6b74vbxhbgabymjje4ugq

Image Reconstruction with Predictive Filter Flow [article]

Shu Kong, Charless Fowlkes
2018 arXiv   pre-print
We propose a simple, interpretable framework for solving a wide range of image reconstruction problems such as denoising and deconvolution.  ...  We test this model on three tasks: non-uniform motion blur removal, lossy-compression artifact reduction and single image super resolution.  ...  Fig. 1 provides an illustration of our overall framework.  ... 
arXiv:1811.11482v1 fatcat:54amuzi33fbvnowjd5twxptnbm

Super-resolution for low quality thumbnail images

Zhiwei Xiong, Xiaoyan Sun, Feng Wu
2008 2008 IEEE International Conference on Multimedia and Expo  
thumbnails, and then use learning-based pair matching to further enhance the high-frequency details in the upsampled images.  ...  To obtain visually pleasurable highresolution versions for this kind of low-resolution images, we first adopt a PDE-based image regularization technique to alleviate the compression noise in the distorted  ...  FRAMEWORK OF OUR SOLUTION The framework of our SR scheme is shown in Figure 2 . An original image Y 0 is first downsampled by a low-pass filter to form a LR version X 0 .  ... 
doi:10.1109/icme.2008.4607401 dblp:conf/icmcs/XiongSW08 fatcat:r2had2zq3zgu3ejrio4sytsy4q
« Previous Showing results 1 — 15 out of 51,706 results