Research map / Video / Video-language understanding
Measuring the Importance of Temporal Features in Video Saliency
Matthias Tangemann, Matthias Kümmerer, Thomas S.A. Wallis, Matthias Bethge
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 |
|---|---|---|---|
| average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame prediction | DeepGaze II · LEDOV val | information gain (bit): 1.216 -> 1.445 | Improved |
| average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame prediction | DeepGaze II · LEDOV test | information gain (bit): 1.117 -> 1.367 | Improved |
| average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame prediction | DeepGaze II · DIEM | information gain (bit): 0.674 -> 0.660 | Hurt |
| average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame prediction | DeepGaze II · Meta-benchmark (LEDOV & DIEM) | information gain (bit): 0.787 -> 0.528 | Hurt |
| static model DeepGaze MR versus temporal video model TASED-Net | DeepGaze MR · LEDOV val | NSS: 3.498 -> 3.857 | Improved |
| static model DeepGaze MR versus temporal video model TASED-Net | DeepGaze MR · DIEM | NSS: 3.194 -> 3.116 | Hurt |
| static model DeepGaze MR versus temporal video model TASED-Net | DeepGaze MR · Meta-benchmark (LEDOV & DIEM) | NSS: 2.855 -> 2.458 | Hurt |
| static features explain share of explainable information on LEDOV | DeepGaze MR · LEDOV val | percent explainable information: 62.8 -> 74.6 | Improved |
| select window length 16 frames by grid search (ablation) | DeepGaze MR · LEDOV | information gain | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- TASED-Net — TASED-Net: Temporally-Aggregating Spatial Encoder-Decoder Network for Video Saliency Detection (ICCV 2019) · better on 3 of 5 shared results
- DeepVS — DeepVS: A Deep Learning Based Video Saliency Prediction Approach (ECCV 2018) · better on 2 of 2 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
- Revisiting Video Saliency: A Large-Scale Benchmark and a New Model (CVPR 2018 · similarity 0.92)
- Revisiting Video Saliency Prediction in the Deep Learning Era (TPAMI 2019 · similarity 0.91)
- Saliency Prediction in the Deep Learning Era: Successes and Limitations (TPAMI 2019 · similarity 0.90)
- A Motion-aware Spatio-temporal Graph for Video Salient Object Ranking (NeurIPS 2024 · similarity 0.89)
- CaRDiff: Video Salient Object Ranking Chain of Thought Reasoning for Saliency Prediction with Diffusion (AAAI 2025 · similarity 0.86)
- StreamGaze: Gaze-Guided Temporal Reasoning and Proactive Understanding in Streaming Videos (NeurIPS 2026 · similarity 0.86)
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.
