mlsanityData to 14 Oct 2026

Hand-Shadow Poser

arXiv 2505.07012 ↗First posted 11 May 2025cs.CG cs.AIcited by 4

Not on the map. Why?

Abstract (arXiv)

Hand shadow art is a captivating art form, creatively using hand shadows to reproduce expressive shapes on the wall. In this work, we study an inverse problem: given a target shape, find the poses of left and right hands that together best produce a shadow resembling the input. This problem is nontrivial, since the design space of 3D hand poses is huge while being restrictive due to anatomical constraints. Also, we need to attend to the input's shape and crucial features, though the input is colorless and textureless. To meet these challenges, we design Hand-Shadow Poser, a three-stage pipeline, to decouple the anatomical constraints (by hand) and semantic constraints (by shadow shape): (i) a generative hand assignment module to explore diverse but reasonable left/right-hand shape hypotheses; (ii) a generalized hand-shadow alignment module to infer coarse hand poses with a similarity-driven strategy for selecting hypotheses; and (iii) a shadow-feature-aware refinement module to optimize the hand poses for physical plausibility and shadow feature preservation. Further, we design our pipeline to be trainable on generic public hand data, thus avoiding the need for any specialized training dataset. For method validation, we build a benchmark of 210 diverse shadow shapes of varying complexity and a comprehensive set of metrics, including a novel DINOv2-based evaluation metric. Through extensive comparisons with multiple baselines and user studies, our approach is demonstrated to effectively generate bimanual hand poses for a large variety of hand shapes for over 85% of the benchmark cases.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 46 of its 113 references that are in the map. First posted on arXiv 11 May 2025.

Builds on

46 of 113 references.

References · 46

  1. 2017Attention Is All You Needcited by 60,361
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  5. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  6. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  7. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  8. 2020Denoising Diffusion Implicit Modelscited by 9,781
  9. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  10. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  11. 2019Expressive Body Capture: 3D Hands, Face, and Body from a Single Imagecited by 1,612
  12. 2020Meta-Learning in Neural Networks: A Surveycited by 950
34 more
  1. 2019Meta-Learning with Implicit Gradientscited by 525
  2. 2023Instruct-NeRF2NeRF: Editing 3D Scenes with Instructionscited by 484
  3. 2021DexYCB: A Benchmark for Capturing Hand Grasping of Objectscited by 325
  4. 2023Reconstructing Hands in 3D with Transformerscited by 306
  5. 2020I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Imagecited by 274
  6. 2019FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Imagescited by 271
  7. 2020InterHand2.6M: A Dataset and Baseline for 3D Interacting Hand Pose Estimation from a Single RGB Imagecited by 245
  8. 2017Learning to Estimate 3D Hand Pose from Single RGB Imagescited by 234
  9. 2017Hand Keypoint Detection in Single Images using Multiview Bootstrappingcited by 231
  10. 2022AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Timecited by 200
  11. 2020Whole-Body Human Pose Estimation in the Wildcited by 178
  12. 2018Monocular Total Capture: Posing Face, Body, and Hands in the Wildcited by 162
  13. 2021Hand-Object Contact Consistency Reasoning for Human Grasps Generationcited by 160
  14. 2020CPF: Learning a Contact Potential Field to Model the Hand-Object Interactioncited by 128
  15. 2019End-to-end Hand Mesh Recovery from a Monocular RGB Imagecited by 124
  16. 2015Capturing Hands in Action using Discriminative Salient Points and Physics Simulationcited by 115
  17. 2018Hand Pose Estimation via Latent 2.5D Heatmap Regressioncited by 111
  18. 2020Monocular Real-time Hand Shape and Motion Capture using Multi-modal Datacited by 106
  19. 2024StyleGaussian: Instant 3D Style Transfer with Gaussian Splattingcited by 79
  20. 2021MobRecon: Mobile-Friendly Hand Mesh Reconstruction from Monocular Imagecited by 74
  21. 2019Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Taskscited by 48
  22. 2023Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Modelscited by 46
  23. 2021Hand Image Understanding via Deep Multi-Task Learningcited by 31
  24. 2024A Simple Baseline for Efficient Hand Mesh Reconstructioncited by 27
  25. 2024Factorized Diffusion: Perceptual Illusions by Noise Decompositioncited by 23
  26. 2017Large-scale Multiview 3D Hand Pose Datasetcited by 20
  27. 2023Wired Perspectives: Multi-View Wire Art Embraces Generative AIcited by 17
  28. 2022Designing Perceptual Puzzles by Differentiating Probabilistic Programscited by 12
  29. 2024LAKE-RED: Camouflaged Images Generation by Latent Background Knowledge Retrieval-Augmented Diffusioncited by 12
  30. 2021Shadow Art Revisited: A Differentiable Rendering Based Approachcited by 10
  31. 2023Constructing Printable Surfaces with View-Dependent Appearancecited by 6
  32. 2024SD-πXL: Generating Low-Resolution Quantized Imagery via Score Distillationcited by 3
  33. 2019Silhouette-Net: 3D Hand Pose Estimation from Silhouettescited by 2
  34. 2015Printed Perforated Lampshades for Continuous Projective Imagescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 3.

4 papers.

  1. 2025ShadowDraw: From Any Object to Shadow-Drawing Compositional Artcited by 1
  2. 2026CNS-Edit++: Category-Agnostic 3D Editing with Coupled Neural Shape Representationcited by 0
  3. 2026Robot Learning to Communicate through Projected Visual Abstractionscited by 0
  4. 2026Differentiable Voxelization of Surface Representationscited by 0