IEEE

CLIP-to-Seg Distillation for Zero-shot Semantic Segmentation

Abstract: CLIP has greatly advanced zero-shot segmentation by leveraging its strong visual-language association and generalization capability. However, directly adapting CLIP for segmentation often ...
GitHub

Data-Segmentation Prompt based Continual Learning Framework for Online Spatio-Temporal Prediction (DepCL)

Dynamic spatio-temporal networks face critical challenges, including catastrophic forgetting of historical patterns and the prohibitive cost of retraining on evolving data. To address these issues, we ...
GitHub

A production-ready implementation of instance segmentation using state-of-the-art models from PyTorch and Hugging Face Transformers.

├── src/ # Source code │ ├── instance_segmentation.py # Core segmentation implementation │ └── data_generator.py # Synthetic data generation ├── web_app/ # Web interface │ └── app.py # Streamlit ...
IEEE

Bonferroni Mean Pre-aggregation Operator Assisted Dynamic Fuzzy Histogram Equalization for Retinal Vascular Segmentation

Abstract: Vasculature feature segmentation from the fundus images is critical for the identification of retinal diseases. However, automated vessel segmentation is challenging owing to variability in ...