Research map / Video / Video segmentation and tracking
LOMo: Latent Ordinal Model for Facial Analysis in Videos
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Model multiple discriminative sub-events and learn their temporal order instead of using MIL | LOMo · CK+ | average class accuracy: 90.8 -> 92.0; 1.2% absolute | Helped |
| Model multiple discriminative sub-events and learn their temporal order instead of using MIL | LOMo · Oulu-CASIA VIS | average accuracy: 69.8 -> 74.0; 4.2% absolute | Helped |
| Model multiple discriminative sub-events and learn their temporal order instead of using MIL | LOMo · UNBC McMaster | classification rate at ROC-EER: 85.9 -> 87.0; 1.1% absolute | Helped |
| Model multiple discriminative sub-events and learn their temporal order instead of MIL | LOMo · LILiR Agree | AUC: 77.7 -> 79.4; 1.7% absolute | Helped |
| Model multiple discriminative sub-events and learn their temporal order instead of MIL | LOMo · LILiR Question | AUC: 80.7 -> 86.6; 5.9% absolute | Helped |
| Use LOMo rather than max temporal pooling | LOMo · LILiR Agree | AUC: 85.5 -> 79.4 | Hurt |
| Use LOMo rather than mean pooling | LOMo · LILiR Understand | AUC: 79.4 -> 80.3 | Helped |
| Use face-recognition CNN features instead of the other tested facial features | facial analysis models · Oulu-CASIA VIS and CK+ | accuracy | Hurt |
| Vary the regularization parameter λ | LOMo · Oulu-CASIA VIS and UNBC McMaster | classification accuracy | No clear effect |
| Vary the number of PCA dimensions | LOMo · Oulu-CASIA VIS and UNBC McMaster | classification accuracy | No clear effect |
Closest papers by meaning
- Learning Facial Action Units From Web Images With Scalable Weakly Supervised Clustering (CVPR 2018 · similarity 0.87)
- Facial Dynamics in Video: Instruction Tuning for Improved Facial Expression Perception and Contextual Awareness (AAAI 2026 · similarity 0.85)
- MART: Masked Affective RepresenTation Learning via Masked Temporal Distribution Distillation (CVPR 2024 · similarity 0.84)
- Facial-R1: Aligning Reasoning and Recognition for Facial Emotion Analysis (AAAI 2026 · similarity 0.84)
- PrefAce: Face-Centric Pretraining with Self-Structure Aware Distillation (AAAI 2024 · similarity 0.84)
- VidEmo: Affective-Tree Reasoning for Emotion-Centric Video Foundation Models (NeurIPS 2025 · similarity 0.83)
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.