pySLAM is a hybrid Python/C++ Visual SLAM pipeline supporting monocular, stereo, and RGB-D cameras. It provides a broad set of modern local and global feature extractors, multiple loop-closure strategies, a volumetric reconstruction module, integrated depth-prediction models, and semantic segmentation capabilities for enhanced scene understanding.

Algorithms for similar image search/reverse image search

3D object detection on the nuScenes dataset

[ICADL] Named entity recognition architecture combining contextual and global features

Global Autoregressive Models (GAMs) for Data-Efficient Sequence Learning

A hybrid CNN–Transformer framework for precise industrial surface defect detection and segmentation, integrating Vision Transformer (ViT) with convolutional modules to effectively capture both local texture details and global contextual features.

Robust image copy detection using deep autoencoder features and moment-based hashing with FAISS for fast large-scale retrieval.

Decision Tree and Global Feature Extract Application