Research map / Video / Video generation
Generic Event Boundary Detection via Denoising Diffusion
Jaejun Hwang, Dayoung Gong, Manjin Kim, Minsu Cho
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 |
|---|---|---|---|
| condition on temporal self-similarity E instead of visual features F (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 68.5 -> 74.0 | Improved |
| condition on temporal self-similarity E instead of visual features F (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | Diversity: 24.3 -> 20.4 | Hurt |
| increase DDIM inference steps from 1 to 32 (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 64.0 -> 74.0 | Improved |
| increase inference steps from 32 to 50 (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 74.0 -> 73.9 | No clear effect |
| use diffusion process instead of non-diffusion variant (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 73.4 -> 74.0 | Improved |
| use diffusion model instead of cVAE (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 62.7 -> 74.0 | Improved |
| swap DDIM for DPM-Solver++ sampler (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 74.0 -> 73.8 | No clear effect |
| swap DDIM for UniPC sampler (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | Diversity: 20.4 -> 17.7 | Hurt |
| increase CFG weight w from 0.6 to 10.0 (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 73.96 -> 71.91 | Hurt |
| increase CFG weight w from 0.1 to 10.0 (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | Diversity: 24.64 -> 10.60 | Hurt |
| increase CFG weight w from 0.1 to 4.0 (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_p2g: 74.24 -> 76.49 | Improved |
| increase annotations N_G from 1 to 4 (ablation) | DiffGEBD · Kinetics-GEBD (diversity-aware) | F1_sym: 70.9 -> 74.0 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Temporal Perceiver — Temporal Perceiver: A General Architecture for Arbitrary Boundary Detection (TPAMI 2023) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DiffDVC: Accurate Event Detection for Dense Video Captioning via Diffusion Models (AAAI 2025 · similarity 0.88)
- Video Anomaly Detection with Motion and Appearance Guided Patch Diffusion Model (AAAI 2025 · similarity 0.87)
- DiffSED: Sound Event Detection with Denoising Diffusion (AAAI 2024 · similarity 0.86)
- Diffusion-based Event Generation for High-Quality Image Deblurring (CVPR 2025 · similarity 0.85)
- The Crystal Ball Hypothesis in diffusion models: Anticipating object positions from initial noise (ICLR 2025 · similarity 0.85)
- Generalized Diffusion Detector: Mining Robust Features from Diffusion Models for Domain-Generalized Detection (CVPR 2025 · similarity 0.84)
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.
