arXiv subjects

Haomian Huang

Publications and source records attributed to Haomian Huang.

4 recordsLinked to original sources

arXiv2026

SignDino: Self-Supervised Sign Language Representation Learning via Temporal-Axis Self-Distillation

Self-supervised sign language representation learning must model two properties not central to natural-image SSL: signs are produced by a small set of anatomically distinct articulators, and their meaning depends on the temporal organisation of those articulators. We introduce SignDino, a self-supervised sign-video encoder that moves the DINOv3 student--teacher recipe from the spatial domain of image crops to the temporal domain of tracked sign streams. Each video is decomposed into left-hand, right-hand, and face streams by a detector-first YOLOv8n+ByteTrack pipeline. A frozen DINOv3 ViT-B/16 embeds each per-frame anatomical crop, while lightweight temporal Transformers, not the image backbone, form the student and EMA teacher. They are trained by temporal DINO self-distillation, frame-level masked-token prediction in the style of iBOT, KoLeo feature spreading, and Gram anchoring of the frame-to-frame similarity structure. This design keeps strong image-level visual primitives fixed and learns only how articulator states evolve across time. We evaluate on sign-to-English translation, isolated sign recognition, and fingerspelling detection benchmarks. Across these tasks, SignDino provides a strong public self-supervised representation and shows competitive or state-of-the-art performance under matched downstream evaluation.

cs.CV↗

arXiv2026

VTaMo: Video-Text Alignment Model for Sign Language Translation

Sign language translation (SLT) converts continuous sign videos into spoken language text. Gloss-free approaches leverage pre-trained visual encoders and language models but rely on implicit cross-modal alignment from translation supervision alone. We present VTaMo, a framework that introduces explicit multi-granularity alignment at three levels: (1) local alignment via entropy-regularized optimal transport with a learnable null token for fine-grained frame-to-token correspondences; (2) global alignment via a learnable orthogonal transformation that calibrates embedding space geometry through Earth Mover's Distance; and (3) position-aligned contrastive learning for discriminative token-level representations. Experiments on Phoenix-2014T, CSL-Daily, How2Sign, and OpenASL demonstrate consistent state-of-the-art performance, with ablations confirming the complementary contributions of each component. Code is available at https://github.com/junyi2005/vtamo.

cs.CV↗

arXiv2026

SignNet-1M: Large-Scale Multilingual Sign Language Video Dataset with Downstream Benchmarks

Sign language models are typically trained on datasets captured under constrained conditions, with limited viewpoint, background, and signer-identity diversity, leading to poor robustness under real-world distribution shifts. We introduce SignNet-1M, a large-scale augmented dataset spanning ASL, CSL, and German Sign Language (DGS). SignNet-1M synthesizes realistic variations along three axes: (i) novel-view rendering (rotation and zoom) via 3D Gaussian Splatting (3DGS), (ii) scene/identity editing via diffusion models for background replacement and signer substitution while preserving sign motion and linguistic content, and (iii) post-rendering augmentations that emulate capture and compression artifacts (e.g., pose/temporal perturbations and video-level corruptions) to better match in-the-wild recordings. Beyond data release, we provide a unified benchmark suite across downstream tasks (e.g., translation and recognition) and ablations that isolate each augmentation component. Experiments across backbones show that training with SignNet-1M consistently improves generalization under cross-view, cross-background, cross-identity, and post-rendering shifts, while maintaining strong in-distribution performance. The dataset, full augmentation pipeline, and benchmark are available at https://signnet.chatsign.ai/.

cs.CV↗

arXiv2023

B-rep Boolean Resulting Model Repair by Correcting Intersection Edges Based on Inference Procedure

As the most essential part of CAD modeling operations, boolean operations on B-rep CAD models often suffer from errors. Errors caused by geometric precision or numerical uncertainty are hard to eliminate. They will reduce the reliability of boolean operations and damage the integrity of the resulting models. And it is difficult to repair false boolean resulting models damaged by errors. In practice, we find that the illegal boolean resulting models stem from the false intersection edges caused by errors. Therefore, this paper proposes an automatic method based on set reasoning to repair flawed structures of the boolean resulting models by correcting their topological intersection edges. We provide a local adaptive tolerance estimation method for each intersection edge based on its geometric features as well as its origin. Then, we propose a set of inference mechanisms based on set operations to infer whether a repair is needed based on the tolerance value and how to correct the inaccurate intersection edge. Our inference strategies are strictly proven, ensuring the reliability and robustness of the repair process. The inference process will transform the problem into a geometric equivalent form less susceptible to errors to get a more accurate intersection edge. Since our inference procedure focuses on topological features, our method can repair the flawed boolean resulting models, no matter what source of errors causes the problem.

cs.GR↗