Learn to implement object detection and bounding box capabilities using Florence 2 vision models in this comprehensive tutorial video. Master the Florence 2 architecture, understand the mechanics of bounding boxes and segmentation, and gain hands-on experience with Google Colab implementation. Follow step-by-step instructions for annotating custom datasets with bounding boxes, fine-tuning Florence 2 models, and setting up LoRA adapters. Explore practical aspects like selecting optimal learning rates and training arguments, while evaluating fine-tuning results and performance metrics. Access complete resources including Colab notebooks, datasets, research papers, and code templates to enhance your computer vision development skills.
Object and Bounding Box Detection with Vision Models Using Florence 2
Overview
Syllabus
Object and Bounding Box Detection with Vision Models
Video Overview
Florence 2 Architecture
How bounding boxes and segmentation work
Inferencing Florence 2 in Google Colab. Bounding boxes and segmentation.
Annotating a custom dataset with bounding boxes.
Fine-tuning Florence 2 on bounding boxes. Notebook demo.
Setting up LoRA adapters
How to choose learning rate and training arguments
Fine-tuning results and performance
Video Resources
Taught by
Trelis Research
Related Courses
-
Computer Vision: YOLO Custom Object Detection with Colab GPU
-
Fine-tuning Llama 2 for Tone or Style Using Shakespeare Dataset
-
Tiny Text and Vision Models - Fine-Tuning and API Setup
-
Fine-tuning Optimizations - DoRA, NEFT, LoRA+, and Unsloth
-
Full Fine-Tuning vs LoRA and QLoRA - Comparison and Best Practices
-
Parameter Efficient Fine-Tuning with Multiple LoRA Adapters for Large Language Models
