Registration-Free Learnable Multi-View Capture of Faces in Dense Semantic Correspondence

• URL: http://arxiv.org/abs/2605.01450v1
• Date: Sat, 02 May 2026 13:59:29 GMT
• Title: Registration-Free Learnable Multi-View Capture of Faces in Dense Semantic Correspondence
• Abstract summary: MOCHI is a multi-view 3D face prediction framework trained without requiring registered training data.<n>Standard point-to-surface distances induce training instabilities and visual artifacts in registration-free settings.<n>We propose pointmap- and normal-based losses instead, which provide smoother gradients and superior reconstruction fidelity.
• Score: 30.38509036690193
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent frameworks like ToFu and TEMPEH provide an automated alternative to classical registration pipelines by predicting 3D meshes in dense semantic correspondence directly from calibrated multi-view images. However, these learning-based methods rely on the slow, manual registration pipelines they aim to replace for their training supervision. We overcome this limitation with MOCHI (Multi-view Optimizable Correspondence of Heads from Images), a multi-view 3D face prediction framework trained without requiring registered training data. MOCHI eliminates the registration data dependency by enforcing topological consistency through a pseudo-linear inverse kinematic solver. Semantic alignment is guided by dense keypoints from a 2D landmark predictor trained exclusively on synthetic data. Our analysis further reveals that standard point-to-surface distances induce training instabilities and visual artifacts in registration-free settings. We propose pointmap- and normal-based losses instead, which provide smoother gradients and superior reconstruction fidelity. Finally, we introduce a test-time optimization scheme that refines network weights over a few dozen iterations. This approach bridges the gap between feed-forward efficiency and iterative optimization precision, allowing MOCHI to outperform traditional labor-intensive pipelines in both reconstruction accuracy and visual quality. Code and model are public at: https://filby89.github.io/mochi.

Related papers

• GIRL-DETR: Gradient-Isolated Reinforcement Learning for Video Moment Retrieval [10.034595422596562]
  Video Moment Retrieval (VMR) task requires accurately localizing temporal boundaries aligned with natural language queries.<n>GIRL-DETR introduces RL post-training into a lightweight temporal localization framework for the first time.
  arXiv  Detail & Related papers  (2026-05-30T15:40:00Z)
• Learning 3D Reconstruction with Priors in Test Time [38.04707926304637]
  We introduce a test-time framework for multiview Transformers (MVTs) that incorporates priors to improve 3D tasks without retraining or modifying pre-trained image-only networks.<n>Our method consistently improves performance over base MVTs by a large margin.
  arXiv  Detail & Related papers  (2026-04-04T22:10:28Z)
• Tail-Aware Post-Training Quantization for 3D Geometry Models [58.79500829118265]
  Post-Training Quantization (PTQ) enables efficient inference without retraining.<n>PTQ fails to transfer effectively to 3D models due to intricate feature distributions and prohibitive calibration overhead.<n>We propose TAPTQ, a Tail-Aware Post-Training Quantization pipeline for 3D geometric learning.
  arXiv  Detail & Related papers  (2026-02-02T07:21:15Z)
• Learning to Track Any Points from Human Motion [55.831218129679144]
  We propose an automated pipeline to generate pseudo-labeled training data for point tracking.<n>A point tracking model trained on AnthroTAP achieves annotated state-of-the-art performance on the TAP-Vid benchmark.
  arXiv  Detail & Related papers  (2025-07-08T17:59:58Z)
• Neural Semantic Surface Maps [52.61017226479506]
  We present an automated technique for computing a map between two genus-zero shapes, which matches semantically corresponding regions to one another. Our approach can generate semantic surface-to-surface maps, eliminating manual annotations or any 3D training data requirement.
  arXiv  Detail & Related papers  (2023-09-09T16:21:56Z)
• 3DMOTFormer: Graph Transformer for Online 3D Multi-Object Tracking [15.330384668966806]
  State-of-the-art 3D multi-object tracking (MOT) approaches typically rely on non-learned model-based algorithms such as Kalman Filter. We propose 3DMOTFormer, a learned geometry-based 3D MOT framework building upon the transformer architecture. Our approach achieves 71.2% and 68.2% AMOTA on the nuScenes validation and test split, respectively.
  arXiv  Detail & Related papers  (2023-08-12T19:19:58Z)
• Locally Aware Piecewise Transformation Fields for 3D Human Mesh Registration [67.69257782645789]
  We propose piecewise transformation fields that learn 3D translation vectors to map any query point in posed space to its correspond position in rest-pose space. We show that fitting parametric models with poses by our network results in much better registration quality, especially for extreme poses.
  arXiv  Detail & Related papers  (2021-04-16T15:16:09Z)
• Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation [75.59415852802958]
  Shape-My-Face (SMF) is a powerful encoder-decoder architecture based on an improved point cloud encoder, a novel visual attention mechanism, graph convolutional decoders with skip connections, and a specialized mouth model. Our model provides topologically-sound meshes with minimal supervision, offers faster training time, has orders of magnitude fewer trainable parameters, is more robust to noise, and can generalize to previously unseen datasets.
  arXiv  Detail & Related papers  (2020-12-16T20:02:36Z)
• Deep Shells: Unsupervised Shape Correspondence with Optimal Transport [52.646396621449]
  We propose a novel unsupervised learning approach to 3D shape correspondence. We show that the proposed method significantly improves over the state-of-the-art on multiple datasets.
  arXiv  Detail & Related papers  (2020-10-28T22:24:07Z)
• Deep-3DAligner: Unsupervised 3D Point Set Registration Network With Optimizable Latent Vector [15.900382629390297]
  We propose to develop a novel model that integrates the optimization to learning, aiming to address the technical challenges in 3D registration. In addition to the deep transformation decoding network, our framework introduce an optimizable deep underlineSpatial underlineCorrelation underlineRepresentation.
  arXiv  Detail & Related papers  (2020-09-29T22:44:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.