You will get Custom AI Object Detection Model (YOLO/Florence-2)
Project details
Get a custom AI object detection model built specifically for YOUR use case. Unlike generic pre-trained models, I train YOLO v8/v10 on YOUR data for maximum accuracy.
What makes this service different:
• Custom-trained on YOUR specific dataset — not generic models
• GPU-optimized inference pipeline ready for production
• Full source code included — you own everything
• Real-world tested: I built a 50,000px+ manga panel extraction system using the same technology
I work with: product detection, defect inspection, medical imaging, document analysis, character recognition, vehicle counting, and more.
Tech: Python, PyTorch, YOLO, Florence-2, OpenCV, CUDA
Fast delivery, clean code, clear communication. Let's build your detection model!
What makes this service different:
• Custom-trained on YOUR specific dataset — not generic models
• GPU-optimized inference pipeline ready for production
• Full source code included — you own everything
• Real-world tested: I built a 50,000px+ manga panel extraction system using the same technology
I work with: product detection, defect inspection, medical imaging, document analysis, character recognition, vehicle counting, and more.
Tech: Python, PyTorch, YOLO, Florence-2, OpenCV, CUDA
Fast delivery, clean code, clear communication. Let's build your detection model!
Machine Learning Tools
NumPy, OpenCV, pandas, Python, PyTorch, scikit-learn, TensorFlowWhat's included
| Service Tiers |
Starter
$150
|
Standard
$300
|
Advanced
$600
|
|---|---|---|---|
| Delivery Time | 5 days | 7 days | 14 days |
Number of Revisions | 1 | 2 | 3 |
Number of Model Variations | 1 | 2 | 3 |
Number of Scenarios | 1 | 3 | 5 |
Model Validation/Testing | - | ||
Model Documentation | - | ||
Data Source Connectivity | - | - | - |
Source Code |
Optional add-ons
You can add these on the next page.
Additional Revision
+$30
Additional Model Variation
(+ 3 Days)
+$75Frequently asked questions
About Thai
Computer Vision & AI Engineer | YOLO, PyTorch, Python Automation
Da Nang, Vietnam - 2:43 pm local time
🔹 What I do best:
• Custom Object Detection — YOLO (v8/v10), Florence-2, RF-DETR, GroundingDINO
• AI Model Fine-tuning — LoRA/PEFT training on custom datasets with PyTorch + CUDA
• Image Segmentation — SAM (Segment Anything), OpenCV, Canny edge detection
• Python Automation — Web scraping, data extraction, workflow automation
• AI Web Apps — Streamlit dashboards with real-time inference
🔹 What sets me apart:
• I don't just use pre-trained models — I train, fine-tune, and optimize them for YOUR specific use case
• Built a complete manga/comic panel extraction system processing 50,000px+ images with intelligent AI detection
• Experienced with both speed-optimized (YOLO) and accuracy-focused (Florence-2 VLM) approaches
• GPU-optimized pipelines with fp16/bf16 inference for maximum performance
🔹 Tech Stack:
Python, PyTorch, OpenCV, YOLO, Florence-2, HuggingFace Transformers, SAM, GroundingDINO, Streamlit, Selenium, Playwright
✅ Fast communication & clean, well-documented code. Let's discuss your project!
Steps for completing your project
After purchasing the project, send requirements so Thai can start the project.
Delivery time starts when Thai receives requirements from you.
Thai works on your project following the steps below.
Revisions may occur after the delivery date.
Data Review & Model Planning
I review your images, define detection classes, and prepare the training dataset with annotations.
Model Training & Optimization
Train custom YOLO model on your data with GPU. Fine-tune hyperparameters for best accuracy.