AI for Content Creation Workshop
CVPR Virtual Conference Website: HERE
This contains links to all content and live interactions (video, text chat) for June 15th.
The AI for Content Creation workshop (AICCW) at CVPR 2020 brings together researchers in computer vision, machine learning, and AI. Content creation has several important applications ranging from virtual reality, videography, gaming, and even retail and advertising. The recent progress of deep learning and machine learning techniques allowed to turn hours of manual, painstaking content creation work into minutes or seconds of automated work. For instance, generative adversarial networks (GANs) have been used to produce photorealistic images of items such as shoes, bags, and other articles of clothing, interior/industrial design, and even computer games' scenes. Neural networks can create impressive and accurate slow-motion sequences from videos captured at standard frame rates, thus side-stepping the need for specialized and expensive hardware. Style transfer algorithms can convincingly render the content of one image with the style of another, offering unique opportunities for generating additional and more diverse training data---in addition to creating awe-inspiring, artistic images.
Learned priors can also be combined with explicit geometric constraints, allowing for realistic and visually pleasing solutions to traditional problems such as novel view synthesis, in particular for the more complex cases of view extrapolation.
AI for content creation lies at the intersection of the graphics, the computer vision, and the design community. However, researchers and professionals in these fields may not be aware of its full potential and inner workings. As such, the workshop is comprised of two parts: techniques for content creation and applications for content creation. The workshop has three goals:
- To cover some introductory concepts to help interested researchers from other fields get started in this exciting new area.
- To present selected success cases to advertise how deep learning can be used for content creation.
- Our invited designers will talk about the pain points that designers face using content creation tools.
More broadly, we hope that the workshop will serve as a forum to discuss the latest topics in content creation and the challenges that vision and learning researchers can help solve.
- Deqing Sun, Ming-Yu Liu, Lu Jiang, James Tompkin, Weilong Yang, and Kalyan Sunkavalli.
Accepted works (in random order)
Extended abstracts (4 pages)
- Mimicry: Towards the Reproducibility of GAN Research
Kwot Sin Lee (University of Cambridge); Christopher Town (University of Cambridge)
- Object-Centric Image Generation from Layouts
Tristan Sylvain (Mila); Pengchuan Zhang (Microsoft Research AI); Yoshua Bengio (Mila); R Devon Hjelm (Microsoft Research); Shikhar Sharma (Microsoft Research)
- Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs
Sangwoo Mo (KAIST); Minsu Cho (POSTECH); Jinwoo Shin (KAIST)
- SegAttnGAN: Text to Image Generation with Segmentation Attention
Yuchuan Gou (paii-labs.com); Qiancheng Wu (University of California, Berkeley); Minghao Li (paii-labs.com); Bo Gong (paii-labs.com); Mei Han (paii-labs.com)
- Toward High-quality Few-shot Font Generation with Dual Memory
Junbum Cha (Clova AI Research, NAVER Corp.); Sanghyuk Chun (Clova AI Research, NAVER Corp.); Gayoung Lee (Clova AI Research, NAVER Corp.); Bado Lee (Clova AI Research, NAVER Corp.); Seonghyeon Kim (Clova AI Research, NAVER Corp.); Hwalsuk Lee (Clova AI Research, NAVER Corp.)
- Generating Object Stamps
Youssef Alami Mejjati (University of Bath); Zejiang Shen (Brown University); Michael Snower (Brown University); Aaron Gokaslan (Brown University); Oliver Wang (Adobe Systems Inc); James Tompkin (Brown University); Kwang In Kim (UNIST)
- COCO-FUNIT: Few-Shot Unsupervised Image Translation with a Content Conditioned Style Encoder
Kuniaki Saito (Boston University); Kate Saenko (Boston University); Ming-Yu Liu (NVIDIA)
- RetrieveGAN: Image Synthesis via Differentiable Patch Retrieval
Hung-Yu Tseng (University of California, Merced); Hsin-Ying Lee (University of California, Merced); Lu Jiang (Google Research); Weilong Yang (Google Inc.); Ming-Hsuan Yang (University of California at Merced)
- Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis
Yujun Shen (Dept. of IE, CUHK); Ceyuan Yang (Chinese University of Hong Kong); Bolei Zhou (CUHK)
[PDF] [arXiv] [Project Webpage]
Papers (8 pages)
- Conditional Image Generation and Manipulation for User-Specified Content
David Stap (University of Amsterdam); Maartje A ter Hoeve (University of Amsterdam); Sarah Ibrahimi (University of Amsterdam)
- FaR-GAN for One-Shot Face Reenactment
Hanxiang Hao (Purdue University); Sriram Baireddy (Purdue University); Amy R. Reibman (Purdue University); Edward Delp (Purdue University)
- CalliGAN: Style and Structure-aware Chinese Calligraphy Character Generator
Shan-Jean Wu (National Taiwan University); Chih-Yuan Yang (National Taiwan University); Jane Yung-jen Hsu (National Taiwan University)
- Network Fusion for Content Creation with Conditional INNs
Patrick Esser (Heidelberg University); Robin Rombach (Heidelberg University); Bjorn Ommer (Heidelberg University)
- Text-guided Image Manipulation via Local Feature Editing
Tianhao Zhang (Google Research); Lu Jiang (Google Research); Weilong Yang (Google Inc.)
Papers (8 pages)—also in other proceedings
- Learning to Shadow Hand-drawn Sketches
Qingyuan Zheng (University of Maryland, Baltimore County); Zhuoru Li (Project HAT); Adam Bargteil (University of Maryland, Baltimore County)
[arXiv] [Project Webpage]
- SEAN: Image Synthesis with Semantic Region-Adaptive Normalization
Peihao Zhu (KAUST); Rameen Abdal (KAUST); Yipeng Qin (Cardiff University); Peter Wonka (KAUST)
[arXiv] [Project Webpage]
- MixNMatch: Multifactor Disentanglement and Encoding for Conditional Image Generation
Yuheng Li (University of California Davis); Krishna Kumar Singh (University of California Davis); Utkarsh Ojha (University of California, Davis); Yong Jae Lee (University of California, Davis)
[PDF] [arXiv] [Project Webpage]
- Interpreting the Latent Space of GANs for Semantic Face Editing
Yujun Shen (Dept. of IE, CUHK); Bolei Zhou (CUHK)
[arXiv] [Project Webpage] [Extended Abstract PDF]
- StarGAN v2: Diverse Image Synthesis for Multiple Domains
Yunjey Choi (Clova AI Research, NAVER Corp.); Youngjung Uh (Clova AI Research, NAVER Corp.); Jaejun Yoo (EPFL); Jung-Woo Ha (Clova AI Research, NAVER Corp.)
[arXiv] [Project Webpage]
- SieveNet: A Unified Framework for Robust Image-Based Virtual Try-On
Surgan Jandial* (IIT Hyderabad); Ayush Chopra* (Media and Data Science Research Lab, Adobe); Balaji Krishnamurthy* (Media and Data Science Research Lab, Adobe); Kumar Ayush (Stanford University); Mayur Hemani (Media and Data Science Research Lab, Adobe); Abhijeet Halwai (Microsoft Research)
- Image2StyleGAN++: How to Edit the Embedded Images?
Rameen Abdal (KAUST); Peter Wonka (KAUST); Yipeng Qin (Cardiff University)
As CVPR2020 is now virtual due to COVID-19, so too will be the workshop. More details coming soon!
We call for papers (8 pages not including references) and extended abstracts (4 pages not including references) to be showcased in a poster session, and for interactive demos, both for the AI for Content Creation Workshop at CVPR 2020. Authors of accepted papers and extended abstracts will be asked to post their submissions on arXiv. Both papers and extended abstracts are not archival and will not be included in the proceedings of CVPR 2020 (authors should be aware that some conferences consider peer-reviewed works with >4-pages to be in violation of double submission policies, e.g., ECCV). We will accept work in progress, work that has not been published elsewhere, and work that has been recently published elsewhere including at CVPR 2020. In the interests of fostering a free exchange of ideas, we welcome both novel and previously-published work.
Paper submissions are double blind and in the CVPR template.
March 20, 2020 March 31, 2020, 11:59 PST
Acceptance notification: April 18, 2020
Submission Website: https://cmt3.research.microsoft.com/AI4CCW2020
The best paper and the best demo will be acknowledged with a Titan RTX GPU (kindly provided by our sponsors).
We seek contributions on a variety of aspects on content creation, including but not limited to the following areas:
- Generative models for image/video synthesis
- Image/video editing
- Image/video inpainting
- Image/video extrapolation
- Image/video translation
- Style transfer
- Text-to-image creation
This includes domains and applications for content creation:
- Image and video for enthusiast, VFX, architecture, advertisements, art, ...
- 2D/3D graphic design
- Text and typefaces
- Design for documents, Web
- Fashion, garments, and outfits
- Novel applications and datasets