IRL & Online Program
Advanced Computer Vision
Deep dive into object detection, image segmentation, and real-time video analytics using state-of-the-art architectures.
Hands-on Highlights
- Train YOLOv9 for custom object detection
- Implement real-time pose estimation algorithms
- Build an automated defect detection system
- Deploy CV models using NVIDIA TensorRT
Detailed Syllabus
Week 1-2
Modern CNN Architectures
- Deep dive into ResNet, ConvNeXt, and MobileNet
- Advanced image augmentation pipelines
- Loss functions for classification
- Hyperparameter tuning for vision models
Week 3-5
Object Detection (YOLO)
- Understanding bounding boxes, IoU, and mAP
- The YOLO family architecture (YOLOv8/v9)
- Dataset annotation and preparation
- Training custom object detectors
Week 6-8
Image Segmentation & Pose
- Semantic vs Instance Segmentation (Mask R-CNN)
- U-Net for medical and industrial applications
- Human pose estimation algorithms
- Real-time video tracking (DeepSORT)
Week 9-10
Edge Deployment & TensorRT
- Model quantization (INT8, FP16)
- Exporting models to ONNX
- Optimizing inference with NVIDIA TensorRT
- Deploying on Jetson Nano
Target Roles & Career Paths
Computer Vision Engineer
Machine Vision Specialist
Perception Engineer (Robotics/AV)
Deep Learning Researcher
These are the primary roles you will be equipped to apply for upon successful completion of the course and portfolio projects.