Research map / Vision-language and multimodal / Human and embodied modeling
PsyPhy: A Psychophysics Driven Evaluation Framework for Visual Recognition
Brandon RichardWebster, Samuel E. Anthony, Walter J. Scheirer
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 |
|---|---|---|---|
| increase Gaussian blur level in 2AFC | pretrained CNNs · 2AFC natural and rendered scenes | accuracy: accuracy declining below 80% | Hurt |
| decrease object scale to 20% of original | pretrained CNNs · 2AFC rendered scenes | accuracy: ~60% or lower | Hurt |
| move from 2AFC to 1000-way MAFC decision | pretrained CNNs · rendered x-rotation | accuracy: under ~20% | Hurt |
| add more layers (VGG-16 vs VGG-19) or reorder pooling/normalization (AlexNet vs CaffeNet) (ablation) | CNN architecture · 2AFC perturbations | robustness | No clear effect |
| increase contrast | pretrained CNNs · MAFC natural scenes | accuracy | Hurt |
| decrease contrast | pretrained CNNs · MAFC natural scenes | accuracy | No clear effect |
| apply 50% test-time dropout in fc6 and fc7 (ablation) | AlexNet · MAFC 3D rotation | curve variability: maximum difference 12.2% | No clear effect |
| apply test-time dropout to rendered 3D objects vs natural scenes (ablation) | AlexNet · dropout MAFC | variability: 16.5% | Hurt |
| compare humans to VGG-19 at increased brightness | humans · MAFC brightness | accuracy: ~9% worse than VGG-19 | Hurt |
Closest papers by meaning
- Perceptual Scales Predicted by Fisher Information Metrics (ICLR 2024 · similarity 0.85)
- Multidimensional Observer Model and Perceptual Dimensions of Human Image Quality Assessment (NeurIPS 2026 · similarity 0.85)
- Diff-2-in-1: Bridging Generation and Dense Perception with Diffusion Models (ICLR 2025 · similarity 0.85)
- PerceptionRubrics: Calibrating Multimodal Evaluation to Human Perception (ICML 2026 · similarity 0.85)
- Brain decoding: toward real-time reconstruction of visual perception (ICLR 2024 · similarity 0.85)
- Self-Calibrating BCIs: Ranking and Recovery of Mental Targets Without Labels (NeurIPS 2025 · 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.
