Research map / Video / Action recognition
Of Mice and Mates: Automated Classification and Modelling of Mouse Behaviour in Groups Using a Single Model Across Cages
Michael P. J. Camilleri, Rasneer Sonia Bains, Christopher K. I. Williams
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use logistic regression instead of naive Bayes as observability classifier | Observability classifier · ABODE val, WASTE=8% | UNREL count: 470 -> 381 | Improved |
| use LFB backbone instead of STLT for behaviour classifier | Behaviour classifier · ABODE validation | macro-F1: 0.36 -> 0.61 | Improved |
| use LFB backbone instead of STLT | Behaviour classifier · ABODE validation | normalised log-likelihood: -1.04 -> -2.27 | Hurt |
| use LFB backbone instead of STLT | Behaviour classifier · ABODE validation | accuracy: 0.73 -> 0.74 | No clear effect |
| add lighting enhancement techniques to behaviour classifier input (ablation) | STLT and LFB · ABODE | accuracy | No clear effect |
| full ALM versus prior-probability baseline on behaviour | ALM · ABODE test | macro-F1: 0.09 -> 0.54 | Improved |
| full ALM versus prior-probability baseline on observability | ALM · ABODE test | macro-F1: 0.48 -> 0.61 | Improved |
| full ALM versus baseline on observability accuracy | ALM · ABODE test | accuracy: 0.93 -> 0.88 | Hurt |
| full ALM versus baseline on UNREL samples | ALM · ABODE test | UNREL count: 1506 -> 996 | Improved |
| full ALM versus baseline on WASTE samples | ALM · ABODE test | WASTE count: 0 -> 1558 | Hurt |
| use a single global GBM instead of per-cage models (ablation) | GBM · IMADGE adults | normalised log-likelihood: average drop is just 4.8% | Hurt |
| increase number of latent states beyond 7 (ablation) | GBM per-cage model · IMADGE adults | normalised log-likelihood | No clear effect |
Closest papers by meaning
- The Multi-Agent Behavior Dataset: Mouse Dyadic Social Interactions (NeurIPS 2021 · similarity 0.92)
- MICE: Multi-animal Interaction Context Encoder — A Hierarchical Foundation Model for Mouse Behavior (NeurIPS 2026 · similarity 0.92)
- Mouse Lockbox Dataset: Behavior Recognition for Mice Solving Lockboxes (IJCV 2026 · similarity 0.90)
- MABe22: A Multi-Species Multi-Task Benchmark for Learned Representations of Behavior (ICML 2023 · similarity 0.86)
- BehaVERT: A Transformer-Based Motion Language Model for Decoding Behavioral Semantics in Mice (IJCV 2026 · similarity 0.85)
- MammalNet: A Large-Scale Video Benchmark for Mammal Recognition and Behavior Understanding (CVPR 2023 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
