The paper presents a method for recognizing sequences of key postures and motions in Bharatanatyam Adavus using a CNN for key postures (99% accuracy) and an SVM for motions (84% accuracy). It then matches these sequences to ground truth with an Edit Distance algorithm, achieving 98% accuracy. The study claims novelty by considering both postures and motions, measuring prediction time, and testing scalability across all 15 Adavus and 58 variations.
arXiv:2610.33130v1 Announce Type: new Abstract: Human activity recognition benchmarks underrepresent culturally situated forms of movement and interaction. We introduce Ampe-DB, a multimodal dataset...
arXiv:2607. 13216v1 Announce Type: cross Abstract: Humans recognize movements effortlessly, even from noisy and complex visual input.
arXiv:2609.13269v1 Announce Type: cross Abstract: Gesture recognition on video is normally posed as classification: label each frame, then act on the label. That is adequate for control, where a comm...
The paper introduces Maswao, a dataset of South African Sign Language (SASL) created through automated collection and preparation of video content from public sources, minimal human labor, and weak sentence alignment using audio-to-text ASR and motion analysis. The authors benchmark the dataset on sign‑to‑text translation, showing Transformer models outperform recurrent baselines with BLEU‑1 to BLEU‑4 scores ranging from 13.78 to 1.54. The study highlights the potential of automated dataset creation for under‑represented sign languages and demonstrates promising translation results that encourage further research.
arXiv:2610.14680v1 Announce Type: new Abstract: Sign Language Recognition (SLR) presents complex challenges due to variable signing speeds and spatial intricacies. Specifically, Indian Sign Language...
arXiv:2608. 09400v1 Announce Type: cross Abstract: Research regarding the sign language recognition mostly relies on RGB images, whileas sign language datasets that provide depth images are limited.
The paper introduces Maswao, a new dataset for South African Sign Language (SASL) created through automated collection and preparation of video content from public sources. The authors convert accompanying audio to text, crop sign regions, and use gradient descent to segment sentences, producing weakly sentence-aligned parallel corpora. They benchmark this dataset on sign-to-text translation, showing Transformer models outperform recurrent baselines with BLEU-1 to BLEU-4 scores up to 13.78, 5.21, 2.52, and 1.54, respectively.
arXiv:2608.29123v1 Announce Type: new Abstract: Training a video-generation model from scratch is hard for reasons that precede model design. The feedback loop is long: a failure that appears only af...
arXiv:2609.12993v1 Announce Type: cross Abstract: We investigate the effect of pretrained T5 model scale and explicit motion features on pose-to-text Indian Sign Language Translation (SLT) for the WS...
MudraGen is a conditional diffusion framework that generates realistic RGB images of Samyukta Hasta mudras, the two‑hand gestures used in Bharatanatyam. It introduces geometry‑aware supervision through three objectives—Keypoint Loss, Joint Offset Loss, and Shape Consistency—to ensure anatomical validity, inter‑hand coordination, and cultural nuance. The model outperforms existing generative approaches in visual realism, anatomical correctness, and fine hand‑pose structure, enabling faithful reproduction of complex mudras.
SignDEG (DINO Eats Geometry) is a temporal sign language encoder that incorporates articulated hand geometry as a pre‑training target while using only video at inference. It leverages frozen DINOv3 features of tracked hands and faces, feeding them into per‑stream temporal Transformers, and introduces two geometric objectives—Geometric Latent Prediction (GLP) and Geometric Relation Transfer (GRT)—to supervise intermediate and final blocks. Jointly trained with temporal self‑distillation, SignDEG achieves higher BLEU‑4 scores on How2Sign/OpenASL and improves recall on ASL Citizen and Sem‑Lex datasets compared to SHuBERT, demonstrating that articulated structure can be transferred into an appearance encoder without requiring a hand reconstruction model at deployment.
The paper argues that in structured image classification, the key question is what information a classifier should receive before making a prediction, rather than which algorithm performs best. It proposes a systematic framework for constructing landmark-derived representations—such as coordinate, distance, angle, and hybrid features—and evaluates them on static hand gesture recognition. Experiments show that hybrid representations, which combine complementary geometric components, outperform raw coordinate features and other single-type representations, highlighting the importance of thoughtful feature construction.