This workshop is for the 2nd installment of the AI for Creative Video Editing and Understanding (CVEU), which follows its success on the first launch at ICCV2021.
The workshop brings together researchers, artists and entrepreneurs working on computer vision, machine learning, computer graphics, human computer interaction, and cognitive research.
It aims to bring awareness of recent advances in machine learning technologies to enable assisted creative-video creation and understanding.
Discuss recent advances in creative video understanding, creation and editing. Some of the topics that we plan to discuss are:
2022-10-16 08:00 AM
|Introduction||08:00 AM - 08:10 AM||-|
|Academia Keynote I by Tali Dekel
Layered Neural Atlases for Video Editing
|08:10 AM - 08:40 AM||-|
|Academia Keynote II by Mark Riedl
Computers, Creativity, and Lovelace
|08:40 AM - 09:10 AM||-|
|Industry Keynote I by Misha Tenenbaum
Reimagining Creative Video Editing Education
|09:10 AM - 09:40 AM||-|
|Industry Keynote II by Pinar Seyhan Demirda
Integration of Generative A.I. Models into Film Production
|09:40 AM - 10:10 AM||-|
|Roundtable I Connecting Academia and Industry
||10:10 AM - 11:10 AM||-|
|Oral Paper Presentation
||11:10 AM - 11:50 AM||-|
|Poster Q&A and Break
||11:50 AM - 12:30 PM||-|
|Roundtable II Advanced Image Generation for Video and Art
||12:30 PM - 01:30 PM||-|
|Academia Keynote III by Yagiz Aksoy
Interactive Depth Editing
|01:30 PM - 02:00 PM||-|
|Industry Keynote III by Raffay Hamid
Lights, Camera, AI! – How Prime Video is Inventing AI technologies to Re-define the Production and Consumption of Long-form Video Content
|02:00 PM - 02:30 PM||-|
|Closing Remarks||02:30 PM - 02:40 PM||-|
|STC: Spatio-Temporal Contrastive Learning for Video Instance Segmentation||Paper Video|
|Mitigating Representation Bias in Action Recognition: Algorithms and Benchmarks||Paper Video|
|SegTAD: Precise Temporal Action Detection via Semantic Segmentation||Paper Video|
|Text-Driven Stylization of Video Objects||Paper
|MND: A New Dataset and Benchmark of Movie Scenes Classified by their Narrative Function||Paper
|Are all combinations equal? Combining textual and visual features with multiple space learning for text-based video retrieval||Paper Video|
|Scene-adaptive Temporal Stabilisation for Video Colourisation using Deep Video Priors||Paper Video|
|MOVIE LENS: Discovering and Characterizing Editing Patterns in the Analysis of Short Movie Sequences||Paper
|Extended Abstract Track||Resources|
|CoolGAN: GANs with Transformers Super-Resolving Images||Paper Video|
|Diffusion Models for Video Prediction and Infilling||Paper
|The Same Thing Only Different: Classification of Movies by their Story Types||Paper Video|
|Temporal and Contextual Transformer for Multi-Camera Editing of TV Shows||Paper
|Invited Paper Track||Resources|
|The Anatomy of Video Editing: A Dataset and Benchmark Suite for AI-Assisted Video Editing||Paper
|XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model||Paper Video|
|PrivHAR: Recognizing Human Actions From Privacy-preserving Lens||Paper Video|
|Beyond RGB: Scene-Property Synthesis with Neural Radiance Fields||Paper Video|
$1,000 USD for best paper awards!
This workshop brings together researchers working on computer vision, machine listening, computer graphics, human-computer interaction, and cognitive research. It aims to bring awareness of recent advances in machine learning technologies to enable assisted creative-video creation and understanding. The workshop will include invited talks by experts in the area and give the community opportunities to share their work via oral and poster presentations. We encourage practitioners, designers, students, post-docs, and researchers to submit work describing new ideas, work-in-progress, and previously or concurrently published research. Topics of interest include but are not limited to:
The CVEU workshop welcomes papers on the topics summarized above. We will publish articles in two tracks: 1) In-proceeding Track and and 2) Extended Abstract Track. All the accepted papers must be unpublished and original work related to topics of the CVEU workshop and have opportunities to be presented as orals or posters during the workshop.
The papers are limited to 14 pages excluding references and will be included in the official ECCV workshop proceedings. Please use the ECCV template.
The articles submitted to this track will not be published in conjunction with ECCV 2022 proceedings. So that you could submit it to other conferences or journals. The papers are limited to 4 pages, including figures and tables, formatted according to the ECCV style. Additional pages containing only cited references are allowed. Please use the CVPR template which allows more contents.
All the submitted papers will go through a double-blind peer-review process with no rebuttal or second review cycle. All the paper submissions will be handled by Microsoft CMT. Please submit your paper to our CVEU CMT Console and follow ECCV 2022 Author Instruction.
The Invited Submission Track welcomes published papers on the topics summarized above. We will invite the accpeted papers to be presented during the workshop.
The papers submitted to the Invited Submission Track must be published work (e.g., accepted by ECCV 2022) related to topics of the CVEU workshop. All the submitted papers will have the opportunities to be presented during the workshop.
Please complete this Google Form to submit your paper.