ArticleNature methods2024
Lightning Pose: improved animal pose estimation via semi-supervised learning, Bayesian ensembling and cloud-native open-source tools.
Article in Nature methods, 2024. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 45 papers, 1 of them a synthesis that pooled it.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
45 citing papers in PubMed, 1 synthesis or guideline pooled it.
- Does advancement in marker-less pose-estimation mean more quality research? A systematic review.Frontiers in behavioral neuroscience · 2025Pooled it
- Article
- Article
- Article
- Multi-Camera Analysis of Simulated Porcine Recovery States Based on Three-Dimensional Pose Using Synthetic Data.Journal of imaging · 2026Article
- Using Machine Learning to Automate the Analysis of an Olfactory Habituation-Dishabituation Task in Mice.Brain and behavior · 2026Article
- Cell-type Plasticity Supports Behavioral Adaptations at the Water-to-Land Interface.bioRxiv : the preprint server for biology · 2026Article
- SqueakPose Studio, an end-to-end platform for pose estimation and real-time edge-AI deployment.eLife · 2026Article
- Toward video-LLM driven workflow for behavioral segmentation and scoring in mice performing a skilled water-reaching task: an evaluation of recent LLM models.Neurophotonics · 2026Article
- Chronic Administration of Marinobufagenin in Mice Causes Hyperlocomotion and Decrease in Anxiety by Altering Monoamine Turnover Unaccompanied by Motor Deficits or Oxidative Stress.International journal of molecular sciences · 2026Article
- Article
- Cheese3D enables sensitive detection and analysis of whole-face movement in mice.Nature neuroscience · 2026Article
- Beast3D: Animal behavioral analysis and neural encoding from multi-view video via Gaussian splatting.ArXiv · 2026Article
- NeuroPupil: A generalization-first framework for scalable and biologically informative cross-species pupillometry.bioRxiv : the preprint server for biology · 2026Article
- Unsupervised transfer learning enables multi-animal tracking without training annotation.Nature methods · 2026Article
- Lightning Pose 3D: an uncertainty-aware framework for data-efficient multi-view animal pose estimation.bioRxiv : the preprint server for biology · 2026Article
- Article
- Article
- Animal behavioral analysis and neural encoding with transformer-based self-supervised pretraining.ArXiv · 2026Article
- YORU: Animal behavior detection with object-based approach for real-time closed-loop feedback.Science advances · 2026Article
Corrections and comments
- Update of
Authors and funding
21 authors.
Funding
Abstract
Contemporary pose estimation methods enable precise measurements of behavior via supervised deep learning with hand-labeled video frames. Although effective in many cases, the supervised approach requires extensive labeling and often produces outputs that are unreliable for downstream analyses. Here, we introduce 'Lightning Pose', an efficient pose estimation package with three algorithmic contributions. First, in addition to training on a few labeled video frames, we use many unlabeled videos and penalize the network whenever its predictions violate motion continuity, multiple-view geometry and posture plausibility (semi-supervised learning). Second, we introduce a network architecture that resolves occlusions by predicting pose on any given frame using surrounding unlabeled frames. Third, we refine the pose predictions post hoc by combining ensembling and Kalman smoothing. Together, these components render pose trajectories more accurate and scientifically usable. We released a cloud application that allows users to label data, train networks and process new videos directly from the browser.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.