Meta tags:
keywords=KEYWORD1, KEYWORD2, KEYWORD3, Correspondence, Constraint optimization, Rigid and Deformable Objects Interaction;

Headings (most frequently used words):

deformable, the, rigid, correspondence, with, interactions, for, sparse, dense, manipulation, task, into, meets, guided, robotic, overview, we, stages, keypoints, to, video, our, paper, abstract, method, correspondences, propose, hybrid, based, representation, interaction, tasks, framework, given, demonstration, vlms, decompose, it, first, extract, as, spatial, constraints, then, apply, constraint, optimization, guide, sequence, pretrained, track, on, objects, real, world, results, inserting, hanger, shirt, putting, racket, bag,

Text of the page (most frequently used words):
the (21), deformable (13), rigid (10), for (9), interactions (9), manipulation (8), #sparse (7), and (7), this (6), task (6), with (6), #keypoints (6), dense (6), correspondence (6), interaction (6), page (4), objects (4), are (3), into (3), correspondences (3), hybrid (3), based (3), tasks (3), these (3), information (3), tracking (3), meets (3), guided (3), robotic (3), was (2), project (2), you (2), website (2), academic (2), overview (2), stages (2), first (2), propose (2), representation (2), method (2), such (2), object (2), bodies (2), due (2), rich (2), contacts (2), dynamics (2), therefore (2), representations (2), specific (2), both (2), our (2), built, using, which, adopted, from, free, borrow, source, code, just, ask, that, link, back, footer, licensed, under, creative, commons, attribution, sharealike, international, license, nerfies, template, putting, racket, bag, inserting, hanger, shirt, real, world, results, framework, given, demonstration, vlms, decompose, extract, spatial, constraints, then, apply, constraint, optimization, guide, sequence, pretrained, track, involving, hanging, clothes, dressing, humans, common, daily, life, making, essential, household, robots, compared, single, between, particularly, challenging, multi, point, complex, during, centric, poses, structural, points, without, become, insufficient, work, tailored, capture, intricate, introduce, structure, aware, generated, global, structures, filtered, their, local, however, through, remains, difficult, high, dimensional, further, construct, accurate, keypoint, throughout, design, combines, advantages, encode, fine, grained, while, ensure, efficient, generalization, novel, deformations, shapes, scenarios, together, they, enable, one, shot, transfer, new, minimal, demonstrations, extensive, experiments, demonstrate, effectiveness, broad, applicability, abstract, video, paper, anonymous, authors, research,

Text of the page (random words):
sparse meets dense correspondence guided robotic manipulation with rigid deformable interactions anonymous authors academic research sparse meets dense correspondence guided robotic manipulation with rigid deformable interactions video for our paper sparse meets dense correspondence guided robotic manipulation with rigid deformable interactions abstract manipulation involving rigid deformable interactions such as hanging clothes or dressing humans is common in daily life making it essential for household robots compared to single object manipulation or interactions between rigid bodies these tasks are particularly challenging due to the rich multi point contacts and the complex dynamics of the deformable bodies during interaction therefore object centric representations such as 6d poses or structural points without task specific information become insufficient for these interactions in this work we propose a hybrid correspondence based representation tailored for rigid deformable interactions first to capture intricate interaction information we introduce structure task and interaction aware sparse keypoints the keypoints are generated based on the global structures of both rigid and deformable objects and filtered by their local interaction contacts however tracking these sparse keypoints through the interaction remains difficult due to the high dimensional dynamics of deformable objects therefore we further construct dense correspondences on the deformable objects for accurate keypoint tracking throughout the manipulation this hybrid design combines the advantages of both representations sparse keypoints encode rich task specific information for fine grained manipulation while dense correspondences ensure efficient tracking and generalization to novel deformations shapes and scenarios together they enable one shot transfer to new tasks with minimal demonstrations extensive experiments demonstrate the effectiveness and broad applicability of our method method overview correspondences for rigid deformable interactions we propose a hybrid correspondence based representation for rigid deformable interaction tasks framework overview given a task demonstration vlms decompose it into stages for the stages with rigid deformable interactions we first extract sparse keypoints as spatial constraints then apply constraint optimization to guide the manipulation sequence with pretrained dense correspondence to track keypoints on deformable objects real world results task inserting the hanger into the shirt task putting the racket into the bag this page was built using the academic project page template which was adopted from the nerfies project page you are free to borrow the source code of this website we just ask that you link back to this page in the footer this website is licensed under a creative commons attribution sharealike 4 0 international license