Computer Vision Systems

AI assistant for building face detection, alignment, recognition, and liveness detection systems using ArcFace, AdaFace, and related metric learning frameworks with bias and ethics awareness.

AI assistant for building 2D and 3D human pose estimation systems using MediaPipe, OpenPose, ViTPose, and related frameworks for sports, ergonomics, healthcare, and animation.

AI assistant specialized in semantic, instance, and panoptic segmentation using U-Net, Mask R-CNN, SAM, and vision transformer models for medical and industrial applications.

Specialized AI assistant for developing computer vision models for radiology, pathology, and medical imaging — covering DICOM workflows, segmentation, and regulatory-aware model design.

Expert AI assistant for designing, training, and optimizing object detection models using YOLO, Faster R-CNN, and modern transformer-based architectures.

AI assistant for building and optimizing OCR pipelines, document layout analysis, and intelligent document processing using PaddleOCR, Tesseract, TrOCR, and document AI models.

AI assistant for remote sensing and geospatial computer vision — change detection, land cover segmentation, object detection in satellite imagery using multispectral and SAR data.

AI assistant for depth estimation, 3D scene reconstruction, visual grounding, and multi-modal scene understanding research using NeRF, Gaussian splatting, and vision-language models.

AI assistant for building video analytics pipelines including multi-object tracking, action recognition, crowd counting, and real-time event detection for surveillance and smart city applications.

AI assistant for building automated visual inspection systems for manufacturing defect detection, surface analysis, and quality control using anomaly detection and classification models.