Build, optimize, and deploy image recognition models in production
This advanced training course covers the concepts, architectures, and modern tools of computer vision. It enables participants to design, optimize, and deploy advanced image recognition models based on deep neural networks.
Is it for you ?
Developers, data scientists, AI engineers, and technical experts working on computer vision projects.
Prerequisites
Strong Python skills. Knowledge of Machine Learning and Deep Learning.
What You'll Walk Away With
- ✓ Design advanced computer vision architectures (CNNs, Vision Transformers)
- ✓ Optimize and train image recognition models
- ✓ Configure image processing pipelines for production
- ✓ Evaluate and deploy vision models in real-world environments
Training content
1 Day 1 – Advanced Architectures for Vision
- CNN architectures: ResNet, EfficientNet, MobileNet
- Vision Transformers: ViT and hybrid models
- Fine-tuning and model training
- Preprocessing and data augmentation
Lab / Exercise: Training a fine-tuned CNN model on a real dataset.
2 Key points & takeaways:
- Mastery of CNN architectures and Transformers
- Skills in effective training
3 Day 2 – Production, Optimization, and Deployment
- Vision pipelines: preprocessing, batch inference, GPU optimization
- Quantization, ONNX, real-time optimization
- Deployment: inference API, edge devices, containerization
- Testing, monitoring, drift management
Lab / Exercise: Deploying an optimized model via API + benchmarking.
4 Key points & takeaways:
- Ability to optimize and deploy a model in production
- Comprehensive view of an image recognition pipeline
📌 Practical information
Our training sessions are offered in Montreal or Quebec City, in person or in a virtual classroom. Dates and locations are specified when you select your session below. If you have any questions, check out our FAQ.
