Searcharxiv⌕ Search

arXiv · 2611.18674

SynPanZ: A Synthetic Dataset to Improve Multi-Animal Pose Estimation in Chimpanzees (Pan troglodytes)

Abstract

Pose estimation underpins a wide range of automated monitoring workflows in animal behaviour research. Unfortunately, endangered great apes such as chimpanzees suffer a chronic shortage of the type of hand-annotated training data that has made robust pose estimation in humans and other model organisms possible. This paper introduces SynPanZ, a synthetically generated dataset for pose estimation in great apes. Our pipeline renders multi-individual scenes from rigged, behaviourally animated chimpanzee models with randomised poses, lighting conditions, environmental contexts, and camera motion, producing animated RGB videos with per-instance 2D keypoint locations without manual labelling. To test whether synthetic augmentation improves cross-dataset generalisation, four DeepLabCut models are trained under matched settings on data sourced from DeepWild with varying real-to-synthetic augmentation ratios. Models are evaluated on ChimpACT, a visually disjoint zoo dataset. We find that augmenting the DeepWild training data with approximately three times as many synthetic frames nearly doubles instance coverage (12.34% to 22.54%) while maintaining keypoint localisation accuracy on detected individuals (PCK@0.20: 49.71% to 52.25%). Synthetic-only training fails, showing that rendered data complements rather than replaces real annotated data. This work represents a step towards reliable pose estimation for great apes.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Flora Talyigas, Lindsay Murray, Sonya Hill, Richard Stocker. 2026-10-28. SynPanZ: A Synthetic Dataset to Improve Multi-Animal Pose Estimation in Chimpanzees (Pan troglodytes). https://searcharxiv.com/abs/2611.18674

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

RTime-QA: A Benchmark for Atomic Temporal Event Understanding in Multimodal Large Language Models

Multimodal large language models (MLLMs) perform well on video benchmarks, yet many video questions can be answered from a single frame or from language priors, which leaves open whether these models perceive how an event unfolds in time. We introduce \rtqa, a benchmark for atomic temporal event understanding built on time-reversed question pairs: each video is accompanied by its reversal, and each question is asked on both videos with identical text and options but different correct answers, so that a deterministic model that ignores the order of the frames cannot answer both. \rtqa{} comprises 8,420 questions of five types on 1,278 videos, 639 videos and their reversals. The questions are derived from event annotations that a proposer model writes, a reviewer model checks and human annotators verify, and three subsets target fast evaluation (\rtqatiny), questions that are hard to answer from the first and last frames (\rtqacore) and two consecutive events (\rtqacomp). All 20 evaluated models are less accurate on reversed videos. On \rtqafull, proprietary models answer both questions of 90.6--94.4\% of the pairs and the strongest open-source model of 66.8\%; on \rtqacore, the proprietary models drop to 55.6--64.5\%, and shuffling the frames reduces the paired accuracy of the tested open-source models to 3.4--7.4\%. We further build \rtqatrain{} with the same pipeline, without human verification, and train three open-source MLLMs on it with supervised fine-tuning (SFT) and reinforcement learning (DAPO) under matched settings: both raise paired accuracy on \rtqafull, SFT by 20.7--27.4 points and DAPO by 5.9--22.9 points. For Qwen3.5-9B-Thinking, the training data also help on two other temporal benchmarks: SFT and DAPO raise the Vinoground text score by 5.8 and 3.8 points, respectively, and DAPO raises the TVBench macro accuracy by 4.6 points.

cs.CV↗

Morphology-driven single-domain generalization for cross-cohort Alzheimer's disease detection using 3D MRIs

Although Alzheimer's disease detection via MRIs has advanced significantly thanks to contemporary deep learning models, challenges such as class imbalance, protocol variations, and limited dataset diversity often hinder their generalization capacity. To address this issue, this article focuses on the single domain generalization setting, where given the data of one domain, a model is designed and developed with maximal performance w.r.t. an unseen domain of distinct distribution. Since brain morphology is known to play a crucial role in Alzheimer's diagnosis, we propose the use of learnable pseudo-morphological modules aimed at producing shape-aware, anatomically meaningful class-specific augmentations in combination with a supervised contrastive learning module to extract robust class-specific representations. Experiments conducted across three datasets show improved performance and generalization capacity, especially under class imbalance and imaging protocol variations. The source code will be made available upon acceptance at https://github.com/zobia111/SDG-Alzheimer.

cs.CV↗

Evaluating and Mitigating Skin Color Bias in Skin Lesion Classificatio

Skin lesion classifiers may overlook performance disparity because of subtle skin tone variations by treating skin tone as a categorical attribute. This study proposes a method that represents skin tone as a continuous attribute and assigns individual weights to each sample based on its nonlinear relationship with classification performance. Evaluations on two dermatological datasets showed that fairness metrics were sensitive to skin-color categorization. The proposed method also demonstrated more stable fairness improvements than conventional categorical reweighting while generally preserving classification performance.

cs.CV↗