Detailed Academic Poster Caption: The image displays an academic poster titled "UniGarmentManip: A Unified Framework for Category-Level Garment Manipulation via Dense Visual Correspondence," presented at the Computer Vision and Pattern Recognition (CVPR) conference held in Seattle, WA from June 17-21, 2023. The poster, authored by Ruihai Wu, Haoran Lu, Yiyan Wang, Yubo Wang, and Hao Dong, showcases a comprehensive framework for manipulating garments by leveraging dense visual correspondence techniques. The top left section visualizes the motivation behind the study, emphasizing the variety and complexity of garments in daily life. Examples include different shapes and deformations in tops, trousers, and dresses. The central part of the poster provides an in-depth explanation of "Dense Correspondence for Garments", using detailed diagrams to illustrate the cross-object fabric deformations, including folding and hanging garments. The bottom left portion delves into "Rigid Skeletons for Garments", depicting the transition from rigid objects to flat garments, and further to garments in various states of deformation. The bottom right segment features "One-Shot Manipulation via Dense Correspondence", explaining the process of applying dense visual correspondence to manipulate garments with a single demonstration, including actions like folding and placing. On the right side, the poster presents "Real-World Experiments" with images of robotic setups manipulating real garments, confirming the practical implementation of the framework. Additionally, there are QR codes linked to the project details, Github code repository, and information about postdoc positions sought by the authors. Overall, the poster is a visual summary of the innovative framework and real-world application of garment manipulation technology, highlighting significant advancements in the field of computer vision and robotics. Text transcribed from the image: KING UNIVE UniGarmentManip: A Unified Framework for Category-Level Garment Manipulation via Dense Visual Correspondence Ruihai Wu*, Haoran Lu*, Yiyan Wang, Yubo Wang, Hao Dong (* denotes equal contribution) Motivation: Diverse Garments in Daily Life Dense Correspondence for Garments Self Play PointNet++ Cross Deform Encoder Decoder One/Few-shot Demonstration. Demo shirt Fold Garments CVPR SEATTLE, WA JUNE 17-21, 2024 Real-World Experiments RGB PC sdor Dresses Trousers keleton Shapes r Garments Cross Object & Deform Cross object InfoNCE Loss Deformations Structural Points Skeleton Positive pair ----Negative pair Coarse-to-fine refinement Rigid Objects Garments (Flat) Garments (Cross Deformation) Correspondence Cross Object Policies Hang up clothes Fling One-Shot Manipulation via Dense Correspondence Novel garment Real-world Execution Links ORGO Correspondence Manipulation Polis Project Code Fling Fold Demonstration manipulation Demonstration Demonstration Garment manipulation Real world garment Ruihai Wu Haoran Lu Seeking Postdoc Seeking Ph.D. Position Position 165