I build intelligent, interpretable, and efficient AI systems that combine deep learning, computer vision, time-series forecasting, and signal processing — transforming data into insight and impact.
- 🔬 Researching AI for Healthcare, Explainable Multimodal AI, and Time-Series Forecasting
- ⚙️ Experienced in Deep Learning, Transformer Architectures, and Reinforcement Learning.
- 💡 Passionate about bridging research and practical applications in Computer Vision.
- 🌍 Based in Taiwan | 🌱 Constantly exploring PyTorch, Vision Transformers, and MLOps
- 📫 Reach me via: rasoulameri.github.io/
| Name | Description | Repository |
|---|---|---|
| CV_1 | SegAgent-AI: A Dockerized AI Platform for Industrial Defect Segmentation, Severity Analysis, and Cloud Deployment | Repo Link |
| CV_2 | UNet-based segmentation for detecting carina and ETT tip placement in medical images. | Repo Link |
| CV_3 | Hybrid CNN–Transformer (ViT) framework for industrial surface defect detection and segmentation. | Repo Link |
| CV_4 | Zero-shot language-guided driving perception with Grounding DINO, CLIP, and Qwen2.5-VL. | Repo Link |
| Name | Description | Repository |
|---|---|---|
| ML_1 | AI programming and XAI with Python: data cleaning, model training, and SHAP-based explainability. | Repo Link |
| Name | Description | Repository |
|---|---|---|
| DL_1 | CNN-based digit classification in PyTorch with Grad-CAM XAI and confusion matrix visualization. | Repo Link |
| Name | Description | Repository |
|---|---|---|
| TSF_1 | Hybrid ML framework for river discharge forecasting using ensemble models and AOA hyperparameter tuning. | Repo Link |
Coming soon...
