Meta tags:
description=GarmentPile: Point-Level Visual Affordance Guided Retrieval and Adaptation for Cluttered Garments Manipulation;
keywords=Robotics Manipulation on Garments;

Headings (most frequently used words):

results, garments, pipeline, garmentpile, point, level, visual, affordance, guided, retrieval, and, adaptation, for, cluttered, manipulation, abstract, video, method, large, foundation, model, fails, in, piled, bibtex, overview, details, simulation, real, world, segment, anything, sam2, chatgpt, 4o,

Text of the page (most frequently used words):
the (41), and (31), point (19), affordance (18), #adaptation (16), garments (16), for (12), #manipulation (11), scene (10), level (9), garment (9), retrieval (8), procedure (8), place (8), cluttered (7), can (7), module (7), score (7), result (6), sofa (6), washing (6), machine (6), whole (6), not (6), video (6), predicts (6), pick (6), basket (5), which (5), framework (5), wang (4), points (4), with (4), your (4), browser (4), does (4), support (4), tag (4), visual (3), guided (3), computer (3), only (3), real (3), world (3), results (3), simulation (3), both (3), actionability (3), complex (3), relations (3), row (3), from (2), ruihai (2), zhu (2), ziyu (2), yuran (2), chen (2), yue (2), jiarui (2), dong (2), hao (2), ieee (2), cvf (2), conference (2), vision (2), pattern (2), recognition (2), cvpr (2), 2025 (2), chatgpt (2), that (2), sam2 (2), piled (2), separated (2), leads (2), grab (2), area (2), all (2), different (2), types (2), into (2), our (2), demonstrates (2), effectiveness (2), downstream (2), geometry (2), structure (2), conditioned (2), supervised (2), lower (2), left (2), right (2), upper (2), pipeline (2), cloud (2), when (2), good (2), enough (2), action (2), per (2), selects (2), highest (2), new (2), overview (2), scenarios (2), while (2), higher (2), thus (2), plausible (2), school (2), pku (2), garmentpile (2), website, template, borrowed, unidoormanip, nerfies, inproceedings, wu_2025_cvpr, author, title, booktitle, proceedings, year, bibtex, judge, stacking, relationship, clothes, rgb, depth, images, there, small, chance, successfully, retrieve, perform, well, mainly, because, perfectly, inability, get, appropriate, just, select, center, means, specific, instead, segmented, part, considered, segment, anything, large, foundation, model, fails, construct, representative, realistic, scenes, built, omniverse, isaac, sim, loading, categories, dress, onesie, glove, hat, scarf, trousers, underpants, skirt, top, 126, clothesnet, environment, task, pointnet, backbone, aggregates, local, global, features, facilitate, incorporating, relation, information, each, trained, learning, details, given, observed, initial, proposes, first, after, executing, receives, generates, robot, retrieves, this, loop, executed, until, are, retrieved, method, poses, significant, challenges, robotics, due, deformable, nature, intricate, unlike, single, require, managing, entanglements, interactions, maintaining, cleanliness, stability, address, these, demands, propose, learn, dense, representation, modeling, space, multi, modal, candidates, novel, designs, awareness, inter, object, additionally, introduce, informed, learned, reorganize, configurations, conducive, over, environments, featuring, diverse, pile, abstract, denotes, simultaneously, reveals, suitable, aware, structures, grasping, edges, other, parts, contacting, floor, retrieving, one, dragging, nearby, entangled, out, avoids, manipulating, leading, such, failures, highly, tangled, may, have, guide, reorganizing, will, exist, code, paper, eecs, cfcs, science,

Text of the page (random words):
garmentpile point level visual affordance guided retrieval and adaptation for cluttered garments manipulation garmentpile point level visual affordance guided retrieval and adaptation for cluttered garments manipulation ruihai wu 1 ziyu zhu 2 1 yuran wang 1 yue chen 1 jiarui wang 1 hao dong 1 1 cfcs school of computer science pku 2 school of eecs pku the ieee cvf conference on computer vision and pattern recognition cvpr 2025 paper video code point level affordance for cluttered garments a higher score denotes the higher actionability for downstream retrieval row 1 per point affordance simultaneously reveals 2 garments suitable for retrieval row 2 it is aware of garment structures grasping edges leads other parts contacting floor and relations retrieving one garment while dragging nearby entangled garments out and thus avoids manipulating on points leading to such failures row 3 and 4 highly tangled garments may not have plausible manipulation points affordance can guide reorganizing the scene and thus garments plausible for manipulation will exist abstract cluttered garments manipulation poses significant challenges in robotics due to the complex deformable nature of garments and intricate garment relations unlike single garment manipulation cluttered scenarios require managing complex garment entanglements and interactions while maintaining garment cleanliness and manipulation stability to address these demands we propose to learn point level affordance the dense representation modeling the complex space and multi modal manipulation candidates with novel designs for the awareness of garment geometry structure and inter object relations additionally we introduce an adaptation module informed by learned affordance to reorganize cluttered garments into configurations conducive to manipulation our framework demonstrates effectiveness over environments featuring diverse garment types and pile scenarios in both simulation and the real world video method pipeline overview framework overview given the observed point cloud the affordance module predicts the initial point level manipulation retrieval affordance score when actionability is not good enough the framework proposes the adaptation pick place action it first predicts per point pick affordance and selects the pick point with the highest score conditioned on which it predicts place affordance and selects the place point after executing adaptation action it receives a new point cloud and generates new affordance when actionability is good enough the robot retrieves on the point with the highest affordance score this loop is executed until all garments are retrieved pipeline details learning framework of retrieval pick and place affordance upper left the affordance module predicts the point level retrieval affordance score for the downstream task upper right pointnet backbone aggregates both local and global features that facilitate incorporating garment geometry structure and relation information for each point lower right the place module which predicts the point level place score conditioned on a pick point for adaptation is supervised by the trained affordance module lower left the pick module which predicts the point level place score for adaptation is supervised by the place module results we construct 3 different types of representative and realistic scenes built on omniverse isaac sim loading 9 categories dress onesie glove hat scarf trousers underpants skirt and top of 126 different garments from clothesnet into the environment our framework demonstrates effectiveness in both simulation and real world washing machine scene sofa scene basket scene simulation results sofa scene whole procedure no adaptation basket scene whole procedure no adaptation your browser does not support the video tag washing machine scene whole procedure adaptation real world results washing machine adaptation procedure sofa adaptation procedure your browser does not support the video tag washing machine scene whole procedure no adaptation your browser does not support the video tag sofa scene whole procedure no adaptation your browser does not support the video tag basket scene whole procedure no adaptation large foundation model fails in piled garments segment anything sam2 sam2 can t perform well mainly because 1 the piled garments can t be separated perfectly which leads to the inability to get the appropriate grab points 2 with separated area we can just select the center of the area as grab point which means only specific points instead of all the segmented part can be considered washing machine result sofa result basket result chatgpt 4o chatgpt 4o can t judge the stacking relationship of clothes only by rgb and depth images so there is only a small chance that it can successfully retrieve garments washing machine result sofa result basket result bibtex inproceedings wu_2025_cvpr author wu ruihai and zhu ziyu and wang yuran and chen yue and wang jiarui and dong hao title point level visual affordance guided retrieval and adaptation for cluttered garments manipulation booktitle proceedings of the ieee cvf conference on computer vision and pattern recognition cvpr year 2025 website template borrowed from nerfies unidoormanip