GIF2Video: Color Dequantization and Temporal Interpolation of GIF images Yang Wang, Haibin Huang, Chuan Wang, Tong He, Jue Wang, Minh Hoai. Stony Brook.

Slides:



Advertisements
Similar presentations
Web Design Vocab 3 PNG, JPG, GIF, MP3, MPEG.
Advertisements

Iframes & Images Using HTML.
IJCAI Wei Zhang, 1 Xiangyang Xue, 2 Jianping Fan, 1 Xiaojing Huang, 1 Bin Wu, 1 Mingjie Liu 1 Fudan University, China; 2 UNCC, USA {weizh,
SOMM: Self Organizing Markov Map for Gesture Recognition Pattern Recognition 2010 Spring Seung-Hyun Lee G. Caridakis et al., Pattern Recognition, Vol.
3D Video Generation and Service Based on a TOF Depth Sensor in MPEG-4 Multimedia Framework IEEE Consumer Electronics Sung-Yeol Kim Ji-Ho Cho Andres Koschan.
Interactive Matting Christoph Rhemann Supervised by: Margrit Gelautz and Carsten Rother.
Images.
An Iterative Optimization Approach for Unified Image Segmentation and Matting Hello everyone, my name is Jue Wang, I’m glad to be here to present our paper.
Goals of Computer Vision To make useful decisions based on sensed images To construct 3D structure from 2D images.
Web Design, 5 th Edition 5 Typography and Images.
Internet-scale Imagery for Graphics and Vision James Hays cs195g Computational Photography Brown University, Spring 2010.
Common file formats  Lesson Objective: Understanding common file formats and their differences.  Learning Outcome:  Describe the type of files which.
Addressing Image Compression Techniques on current Internet Technologies By: Eduardo J. Moreira & Onyeka Ezenwoye CIS-6931 Term Paper.
Types of Graphics  Vector Individual scalable objects defined by mathematical equations  Bitmap Pixels (tiny, single-colored square) in a grid Most.
Bitmap Graphics. Bitmap Basics Bitmap Graphic Bitmap Graphic Paint Software Paint Software.
Graphics & Images What File Format Do I Use?. Graphics & Images …..are visual images presented on some form of media (drawings, print, web, digital video)
Image Coloring. Halftone Halftone is the reprographic technique that simulates continuous tone imagery through the use of dots, varying either in size,
Project Two Adding Web Pages, Links, and Images Define and set a home page Add pages to a Web site Describe Dreamweaver's image accessibility features.
Page Design Issues IWM 14 Information Services for the Web.
Data compression. lossless – looking for unicolor areas or repeating patterns –Run length encoding –Dictionary compressions Lossy – reduction of colors.
CHAPTER 4 THE VISUALIZATION PIPELINE. CONTENTS The focus is on presenting the structure of a complete visualization application, both from a conceptual.
Digital Image Editing Presented by John Hohn. File Formats JPEG – Joint Photographic Experts Group PNP – Portable Network Graphics GIF – Graphic Interchange.
Image File Formats Which one is right for me?. The Only Three Image Formats Your Will Ever Need: Names Jpg “Joint Photographic Experts Group” Png “Portable.
Evaluating Climate Visualization: An Information Visualization Approach -By Mridul Sen 1.
SUPER RESOLUTION USING NEURAL NETS Hila Levi & Eran Amar Weizmann Ins
© 2016 Pearson Education, Inc., Hoboken, NJ. All rights reserved.
Conditional Generative Adversarial Networks
Deep Learning for Dual-Energy X-Ray
Sampling, Quantization, Color Models & Indexed Color
Saliency-guided Video Classification via Adaptively weighted learning
"Digital Media Primer" Yue-Ling Wong, Copyright (c)2013 by Pearson Education, Inc. All rights reserved.
Depth Map Upsampling by Self-Guided Residual Interpolation
Compositional Human Pose Regression
Raster Images CPSC 1030.
2D Drawing Basics 1.
Week 4 – Content Generation
FaceNet A Unified Embedding for Face Recognition and Clustering
Presenter: Hajar Emami
Adversarially Tuned Scene Generation
Saving your Animations
Combining Sketch and Tone for Pencil Drawing Production
Proposed (MoDL-SToRM)
Photoshop: Creating and Preparing Images for the Web
Oral presentation for ACM International Conference on Multimedia, 2014
MPEG-Immersive 3DoF+ Standard Work:
Non-Stationary Texture Synthesis by Adversarial Expansion
Enhancing the Enlargement of Images
Image to Image Translation using GANs
Advisor: Chin-Chen Chang1, 2 Student: Yi-Pei Hsieh2
Iterative Crowd Counting
Predicting Body Movement and Recognizing Actions: an Integrated Framework for Mutual Benefits Boyu Wang and Minh Hoai Stony Brook University Experiments:
Good View Hunting: Learning Photo Composition from Dense View Pairs Zijun Wei1, Jianming Zhang2, Xiaohui Shen2, Zhe Lin2, Radomír Měch2, Minh Hoai1, Dimitris.
Comparison of EET and Rank Pooling on UCF101 (split 1)
Sequence-to-Segments Networks for Segment Detection Zijun Wei1, Boyu Wang1, Minh Hoai1, Jianming Zhang2, Xiaohui Shen3, Zhe Lin2, Radomír Měch2, Dimitris.
Multimedia Graphics.
Eigen-Evolution Dense Trajectory Descriptors
Course Recap and What’s Next?
HALO-FREE DESIGN FOR RETINEX BASED REAL-TIME VIDEO ENHANCEMENT SYSTEM
Abnormally Detection
Implementation of a De-blocking Filter and Optimization in PLX
Introduction to Image Analysis and Processing
Chuan Wang1, Haibin Huang1, Xiaoguang Han2, Jue Wang1
Semi-Supervised Learning
U-Net: Convolutional Network for Segmentation
Learning Compositional Visual Concepts with Mutual Consistency
VERY DEEP CONVOLUTIONAL NETWORKS FOR LARGE-SCALE IMAGE RECOGNITION
Cengizhan Can Phoebe de Nooijer
Deep screen image crop and enhance
End-to-End Speech-Driven Facial Animation with Temporal GANs
Rong Ge, Duke University
Presentation transcript:

GIF2Video: Color Dequantization and Temporal Interpolation of GIF images Yang Wang, Haibin Huang, Chuan Wang, Tong He, Jue Wang, Minh Hoai. Stony Brook University, Megvii Research USA, UCLA. Motivation GIF (Graphics Interchange Format) is a highly portable graphics format that is ubiquitous on the Internet GIFs often have much worse visual quality than their original source videos, due to heavy quantization in the creation process We propose GIF2Video, the first learning-based method for enhancing the visual quality of GIFs in the wild GIF Creation & Artifacts Datasets GIF-Faces: human face centric; GIF-Moments: generic & diverse. Color Dequantization Quantization 𝐺=𝑓 𝒞 𝐼 Dequantization 𝐼=𝑓 𝒞 −1 𝐺 ill-posed problem, as 𝑓 𝒞 is many-to-one mapping, 𝒞 is the color palette We seek image 𝐼: min 𝐼 𝐼 − 𝐼 2 2 +𝜆 𝑓 𝒞 𝐼 −𝐺 2 2 ② is not applicable, because is 0 almost everywhere We propose CCDNet to iteratively optimize for ② CCDNet: Compositional Color Dequantization Network Architecture Loss GIF2Video SuperSlomo [H. Jiang, et al. 2018] Conclusions Our method can dramatically enhance the visual quality of input GIFs and significantly reduce the quantization artifacts It is beneficial to separately process dithered & non-dithered GIFs U-Net is an effective building block for CCDNet It is critical to include the loss defined on the gradient values Using adversarial loss yields more realistic images It is beneficial to unfold CCDNet by multiple steps It is beneficial to embed the quantization process into the CCDNet Qualitative Results ① ②