Research map / Vision-language and multimodal / Human and embodied modeling
Learning Visual Attention to Identify People With Autism Spectrum Disorder
Ming Jiang, Qi Zhao
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 |
|---|---|---|---|
| fine-tune VGG-16 on the selected 100 images (ablation) | SVM on VGG-16 features · ASD classification, 100 images | accuracy: 0.83 -> 0.92 | Improved |
| fine-tune VGG-16 on all 700 images (ablation) | SVM on VGG-16 features · ASD classification, 700 images | accuracy: 0.85 -> 0.85 | No clear effect |
| add Fisher-score image selection (100 of 700) to gaze features (ablation) | SVM on gaze features · ASD classification | sensitivity: 0.83 -> 0.93 | Improved |
| add Fisher-score image selection to gaze features (ablation) | SVM on gaze features · ASD classification | accuracy: 0.81 -> 0.86 | Improved |
| add Fisher-score image selection to fine-tuned VGG-16 (ablation) | SVM on VGG-16 fine-tuned · ASD classification | accuracy: 0.85 -> 0.92 | Improved |
| add Fisher-score image selection to non-fine-tuned VGG-16 (ablation) | SVM on VGG-16 · ASD classification | accuracy: 0.85 -> 0.83 | Hurt |
| fine-tune VGG-16 on selected images (ablation) | SVM on VGG-16 features · ASD classification, 100 images | specificity: 0.84 -> 0.92 | Improved |
| use fine-tuned DNN features instead of gaze features on selected images (ablation) | SVM · ASD classification, 100 images | accuracy: 0.86 -> 0.92 | Improved |
| use proposed method instead of Liu et al. eye tracking | Eye tracking ASD diagnosis · ASD diagnosis comparison | accuracy: 0.89 -> 0.92 | Improved |
| use proposed method instead of ADOS | ASD diagnosis · ASD diagnosis comparison | accuracy: 0.82 -> 0.92 | Improved |
| use Fisher score instead of saliency metrics for image selection (ablation) | Image selection · ASD classification (AUC vs number of images) | AUC | Improved |
Closest papers by meaning
- Attention-Based Autism Spectrum Disorder Screening With Privileged Modality (ICCV 2019 · similarity 0.90)
- A Learning Paradigm for Selecting Few Discriminative Stimuli in Eye-Tracking Research (TPAMI 2025 · similarity 0.90)
- Understanding Unique Behavioral Patterns through Multimodal Analysis of Eye-Hand Coordination in Autistic Children (Student Abstract) (AAAI 2025 · similarity 0.85)
- EyeGraph: Modularity-aware Spatio Temporal Graph Clustering for Continuous Event-based Eye Tracking (NeurIPS 2024 · similarity 0.81)
- DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images (NeurIPS 2025 · similarity 0.80)
- Statistical Test for Attention Maps in Vision Transformers (ICML 2024 · similarity 0.80)
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.
