Dear colleague,
We are happy to announce the starting of our NeurIPS 2020 3D + texture garment reconstruction competition. Sorry for multiple copies. What is the competition about? Humans are important targets in many applications. Accurately tracking, capturing, reconstructing and animating the human body, face and garments in 3D are critical for human-computer interaction, gaming, special effects and virtual reality. In the past, this has required extensive manual animation. Regardless of the advances in human body and face reconstruction, still modeling, learning and analyzing human dynamics need further attention. In this competition we plan to push the research in this direction, e.g. understanding human dynamics in 2D and 3D, with special attention to garments. We provide a large-scale dataset (more than 2M frames) of animated garments with variable topology and type. The dataset contains paired RGB images with 3D garment vertices in a sequence. We paid special care to garment dynamics and realistic rendering of RGB data, including lighting, fabric type and texture. We designed three tracks so participants can compete to develop the best method to perform 3D garment reconstruction and texture estimation in a sequence from (1) 3D garments and (2) RGB images. More details are available here <http://chalearnlap.cvc.uab.es/challenge/40/description/>. Why is it important? There has been a growing interest in the topic recently, both from a research or industrial point of view. However, available datasets have been limited in terms of either the number of samples or garment complexity and topology, body shape, pose and garment dynamics (a comparison of available datasets can be seen here <https://arxiv.org/abs/1912.02792>). A large-scale benchmark dataset has been a demand to better study the topic. This competition provides a great opportunity to study state-of-the-art and further push the research. Details of the dataset can be seen here <http://chalearnlap.cvc.uab.es/dataset/38/description/>. Why is it interesting to participate? - You can test your already available method on our benchmark dataset and compete with other state-of-the-art methods, - You can develop new ideas and submit your paper to NeurIPS Workshop/Competition Proceedings or any conference/journal of your choice, - You can win the competition, receive a certificate and attend NeurIPS to present your work. We provide travel grants to the top winning teams of each track. Also the best student approach will be awarded by one NVIDIA GPU. - All participants are candidates to be invited to write a joint paper along with the organizers. We plan to submit this paper to a top-tier conference/journal. Ready to hack? The schedule is available here <http://chalearnlap.cvc.uab.es/challenge/40/schedule/>. You have around 4 months to finish each track. You can enter the competition by registering at Codalab <https://competitions.codalab.org/>. Each track is accessible by the following links. - Track 1, 3D to 3D garment reconstruction <https://competitions.codalab.org/competitions/24767> - Track 2, RGB to 3D garment reconstruction <https://competitions.codalab.org/competitions/24768> - Track 3, RGB to 3D+Texture garment reconstruction <https://competitions.codalab.org/competitions/24769> NeurIPS competition event In the competition track (Fri, Dec 11 - Sat, Dec 12, detailed schedule will be released later) we will present challenge results and winning participants can present their approaches. We also have a confirmed list of invited speakers who are experts in the field of the competition: - Gerard Pons-Moll, Max Planck Institute for Informatics, - Kristen Grauman, University of Texas at Austin, - Stefanos Zafeiriou, Imperial College London, - Yebin Liu, Tsinghua University. Sponsors ChaLearn <http://www.chalearn.org/>, Facebook Reality Labs <https://research.fb.com/category/augmented-reality-virtual-reality/>, NVIDIA <https://research.nvidia.com/>, Baidu <http://research.baidu.com/> Hope to see you soon in this great event! NeurIPS 2020 ChaLearn organizing team Meysam Madadi, Hugo Bertiche, Wafa Bouzouita, Isabelle Guyon, Sergio Escalera
_______________________________________________ uai mailing list [email protected] https://it.engineering.oregonstate.edu/mailman/listinfo/uai
