SignRefine: Adapting Foundational Video Models for Sign Language Generation

• URL: http://arxiv.org/abs/2609.08496v1
• Date: Tue, 08 Sep 2026 09:38:22 GMT
• Title: SignRefine: Adapting Foundational Video Models for Sign Language Generation
• Abstract summary: SignRefine is a sign language video generation model that produces comprehensible signing from 2D keypoint conditioning alone.<n>We present NVSign, a large-scale dataset of video content produced in sign language.
• Score: 19.287012175897235
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Sign language video generation demands precise hand and facial articulation, yet modern video diffusion models, trained predominantly on spoken-language video, produce artifacts that render signing unintelligible. We propose SignRefine, a sign language video generation model that produces comprehensible signing from 2D keypoint conditioning alone, generalizing across appearances and visual conditions. Our approach builds on a pretrained video diffusion transformer and introduces local adapters with spatial grounding to selectively refine hand and face regions, steering the strong base model's prior toward accurate articulation. To enable this work and support broader sign language research, we present NVSign, a large-scale dataset of video content natively produced in sign language, offering diverse signer appearances, environments, and natural conversational settings. Trained on this data, our model shows up to 30% improvement in hand pose precision metrics over the strongest baseline and is preferred by sign language users for visual quality and comprehensibility in more than 80% of comparisons.

Related papers

• SignVerse-2M: A Two-Million-Clip Pose-Native Universe of 55+ Sign Languages [28.65355856480869]
  We present SignVerse-2M, a large-scale multilingual pose-native dataset for sign language pose modeling and evaluation.<n>It applies DWPose in a unified preprocessing pipeline to convert raw videos into 2D pose sequences that can be used directly for modeling.<n>Unlike many laboratory datasets, this resource preserves the recording conditions and speaker diversity of real-world videos.
  arXiv  Detail & Related papers  (2026-05-03T05:26:26Z)
• Signs as Tokens: A Retrieval-Enhanced Multilingual Sign Language Generator [55.94334001112357]
  We introduce a multilingual sign language model, Signs as Tokens (SOKE), which can generate 3D sign avatars autoregressively from text inputs.<n>We propose a retrieval-enhanced SLG approach, which incorporates external sign dictionaries to provide accurate word-level signs.
  arXiv  Detail & Related papers  (2024-11-26T18:28:09Z)
• SHuBERT: Self-Supervised Sign Language Representation Learning via Multi-Stream Cluster Prediction [65.1590372072555]
  SHuBERT (Sign Hidden-Unit BERT) is a self-supervised contextual representation model learned from 1,000 hours of American Sign Language video.<n>SHuBERT adapts masked token prediction objectives to multi-stream visual sign language input, learning to predict multiple targets corresponding to clustered hand, face, and body pose streams.<n>SHuBERT achieves state-of-the-art performance across multiple tasks including sign language translation, isolated sign language recognition, and fingerspelling detection.
  arXiv  Detail & Related papers  (2024-11-25T03:13:08Z)
• SignMusketeers: An Efficient Multi-Stream Approach for Sign Language Translation at Scale [22.49602248323602]
  A persistent challenge in sign language video processing is how we learn representations of sign language.<n>Our proposed method focuses on just the most relevant parts in a signing video: the face, hands and body pose of the signer.<n>Our approach is based on learning from individual frames (rather than video sequences) and is therefore much more efficient than prior work on sign language pre-training.
  arXiv  Detail & Related papers  (2024-06-11T03:00:41Z)
• Distilling Vision-Language Models on Millions of Videos [62.92789440875999]
  We fine-tune a video-language model from a strong image-language baseline with synthesized instructional data. The resulting video model by video-instruction-tuning (VIIT) is then used to auto-label millions of videos to generate high-quality captions. As a side product, we generate the largest video caption dataset to date.
  arXiv  Detail & Related papers  (2024-01-11T18:59:53Z)
• VideoPoet: A Large Language Model for Zero-Shot Video Generation [78.57171527944774]
  VideoPoet is a language model capable of synthesizing high-quality video with matching audio. VideoPoet employs a decoder-only transformer architecture that processes multimodal inputs.
  arXiv  Detail & Related papers  (2023-12-21T18:46:41Z)
• Pixel Aligned Language Models [94.32841818609914]
  We develop a vision-language model that can take locations as either inputs or outputs. When taking locations as inputs, the model performs location-conditioned captioning, which generates captions for the indicated object or region. Our model is pre-trained on the Localized Narrative dataset, which contains pixel-word-aligned captioning from human attention.
  arXiv  Detail & Related papers  (2023-12-14T18:57:58Z)
• DiffSLVA: Harnessing Diffusion Models for Sign Language Video Anonymization [33.18321022815901]
  We introduce DiffSLVA, a novel methodology for text-guided sign language video anonymization. We develop a specialized module dedicated to capturing facial expressions, which are critical for conveying linguistic information in signed languages. This innovative methodology makes possible, for the first time, sign language video anonymization that could be used for real-world applications.
  arXiv  Detail & Related papers  (2023-11-27T18:26:19Z)
• VideoCon: Robust Video-Language Alignment via Contrast Captions [80.08882631838914]
  Video-language alignment models are not robust to semantically-plausible contrastive changes in the video captions. Our work identifies a broad spectrum of contrast misalignments, such as replacing entities, actions, and flipping event order. Our model sets new state of the art zero-shot performance in temporally-extensive video-language tasks.
  arXiv  Detail & Related papers  (2023-11-15T19:51:57Z)
• Towards Fast Adaptation of Pretrained Contrastive Models for Multi-channel Video-Language Retrieval [70.30052749168013]
  Multi-channel video-language retrieval require models to understand information from different channels. contrastive multimodal models are shown to be highly effective at aligning entities in images/videos and text. There is not a clear way to quickly adapt these two lines to multi-channel video-language retrieval with limited data and resources.
  arXiv  Detail & Related papers  (2022-06-05T01:43:52Z)
• Can Everybody Sign Now? Exploring Sign Language Video Generation from 2D Poses [0.0]
  We use the state of the art in Deep Learning for motion transfer to generate videos of signers performing sign language given a 2D pose skeleton. We evaluate the generated videos quantitatively and qualitatively showing that the current models are not enough to generated adequate videos for Sign Language due to lack of detail in hands.
  arXiv  Detail & Related papers  (2020-12-20T14:43:32Z)
• Everybody Sign Now: Translating Spoken Language to Photo Realistic Sign Language Video [43.45785951443149]
  To be truly understandable by Deaf communities, an automatic Sign Language Production system must generate a photo-realistic signer. We propose SignGAN, the first SLP model to produce photo-realistic continuous sign language videos directly from spoken language. A pose-conditioned human synthesis model is then introduced to generate a photo-realistic sign language video from the skeletal pose sequence.
  arXiv  Detail & Related papers  (2020-11-19T14:31:06Z)

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.