Research map / Video / Action recognition
Feature Hallucination for Self-supervised Action Recognition
Lei Wang, Piotr Koniusz
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 |
|---|---|---|---|
| add Power Normalization (AsinhE, SigmE or AxMin) to the HAF and BoW/FV streams (ablation) | hallucination pipeline · HMDB-51 (split 1) | accuracy: 82.29 -> 83.20 | Improved |
| add multiple sketches per BoW plus Power Normalization (ablation) | HAF*+BoW halluc. (256px) · MPII Cooking Activities | mAP: 77.8 -> 80.4 | Improved |
| add multiple sketches per BoW plus Power Normalization (ablation) | HAF+BoW halluc. (512px) · MPII Cooking Activities | mAP: 78.5 -> 81.7 | Improved |
| double the frame resolution to 512px before human detection | HAF+BoW hal.+MSK/PN · MPII Cooking Activities | mAP: 80.4 -> 81.7 | Improved |
| double the frame resolution to 512px before human detection | HAF+BoW halluc. · MPII Cooking Activities | mAP: 77.8 -> 78.5 | Improved |
Closest papers by meaning
- MASH-VLM: Mitigating Action-Scene Hallucination in Video-LLMs through Disentangled Spatial-Temporal Representations (CVPR 2025 · similarity 0.89)
- MHBench: Demystifying Motion Hallucination in VideoLLMs (AAAI 2025 · similarity 0.87)
- Dual DETRs for Multi-Label Temporal Action Detection (CVPR 2024 · similarity 0.86)
- VidHalluc: Evaluating Temporal Hallucinations in Multimodal Large Language Models for Video Understanding (CVPR 2025 · similarity 0.86)
- Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations (CVPR 2026 · similarity 0.85)
- Analyzing and Mitigating Object Hallucination: A Training Bias Perspective (AAAI 2026 · 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.
