Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Visual Memorability for Robotic Interestingness via Unsupervised Online Learning
Chen Wang, Wenshan Wang, Yuheng Qiu, Yafei Hu, Sebastian Scherer
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 sparse writing protocol with tangent operator (ablation) | Visual memorability interestingness model · SubT Normal (delta=2) | AUC-OP: 0.633 -> 0.842 | Improved |
| add sparse writing protocol with tangent operator (ablation) | Visual memorability interestingness model · SubT Difficult (delta=2) | AUC-OP: 0.373 -> 0.585 | Improved |
| add translational invariance (cross-correlation) to memory reading (ablation) | Visual memorability interestingness model · SubT Normal (delta=2) | AUC-OP: 0.510 -> 0.842 | Improved |
| add translational invariance (cross-correlation) to memory reading (ablation) | Visual memorability interestingness model · SubT Difficult (delta=2) | AUC-OP: 0.268 -> 0.585 | Improved |
| add short-term learning with known uninteresting scenes (ablation) | Visual memorability interestingness model · SubT Normal (delta=2) | AUC-OP: 0.711 -> 0.842 | Improved |
| add short-term learning with known uninteresting scenes (ablation) | Visual memorability interestingness model · SubT Difficult (delta=2) | AUC-OP: 0.450 -> 0.585 | Improved |
| add online learning (ablation) | Visual memorability interestingness model · SubT | AUC-OP: about 20% higher overall performance | Improved |
| use proposed method instead of baseline interestingness method | Visual memorability interestingness model · SubT Normal (delta=1) | AUC-OP: 0.622 -> 0.662 | Improved |
| use proposed method instead of baseline interestingness method | Visual memorability interestingness model · SubT Difficult (delta=1) | AUC-OP: 0.352 -> 0.407 | Improved |
| use proposed method instead of baseline interestingness method | Visual memorability interestingness model · SubT Difficult (delta=4) | AUC-OP: 0.700 -> 0.768 | Improved |
| increase memory capacity (ablation) | Visual memorability interestingness model · random tensor recall | AUC-OP: reading accuracy of first tensor drops less with larger capacity | Improved |
| raise the writing rate gamma_w from 0.2 to 1.0 | Visual memorability interestingness model · Drone Filming | AUC-OP: 1.0 vs 0.2 | Mixed |
Closest papers by meaning
- MemoNav: Working Memory Model for Visual Navigation (CVPR 2024 · similarity 0.85)
- 3D-Mem: 3D Scene Memory for Embodied Exploration and Reasoning (CVPR 2025 · similarity 0.85)
- Kinaema: a recurrent sequence model for memory and pose in motion (NeurIPS 2025 · similarity 0.85)
- Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction (ICLR 2024 · similarity 0.84)
- SAM2Act: Integrating Visual Foundation Model with A Memory Architecture for Robotic Manipulation (ICML 2025 · similarity 0.84)
- PlaySlot: Learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning (ICML 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.
