PPE & Kitchen Hygiene Violation Detection
Budget / Salary$30–250
TypeFreelance project
LocationRemote
Posted2 hours ago
**Job Title:** Computer Vision / Roboflow Expert – Fine-Tune Model for PPE & Kitchen Hygiene Violations
**Project Description:**
We are developing an AI-driven restaurant and kitchen safety monitoring system. We need an experienced Computer Vision & Deep Learning Engineer to build, annotate, and fine-tune a high-accuracy object detection/segmentation model (using Roboflow and YOLO models like YOLOv8/v11) to accurately detect specific hygiene violations in real-time CCTV feeds.
**Key Challenges & Core Requirements:**
1. **Glove Compliance:**
* Detect bare hands vs. gloved hands (handling fine-grained vision tasks on food prep surfaces).
* Accurately flag "No Gloves" violations when staff are touching/preparing food.
2. **Proper Mask Wearing:**
* Detect proper mask usage vs. improper usage (mask pulled down to the chin / nose exposed) vs. no mask.
3. **Hairnet / Head Cover Detection:**
* High-precision detection of kitchen hairnets/caps on cooks, distinguishing between bare head, improper wear, and full coverage under varying lighting and camera angles.
4 Mobile Phone Usage:
Detect employees holding or using smartphones while working/preparing food.
Differentiate between a phone held to the ear, phone held in hand near food prep stations, and hand-only gestures.
**Scope of Work:**
* Review, clean, and augment our existing Roboflow dataset or assist in sourcing/annotating edge-case images.
* Define a robust labeling framework (e.g., multi-class classification or 2-stage object detection: Person $\rightarrow$ Face/Hands $\rightarrow$ Compliance state).
* Train and fine-tune a SOTA detection model (YOLOv8/v11, Roboflow Workflows, or custom PyTorch pipeline).
* Optimize for high Precision and Recall on edge cases (low light, overhead angles, fast hand movement).
* Export model weights (ONNX, PyTorch, or Roboflow Hosted API) for seamless integration into our existing codebase.
**Deliverables:**
* Fully annotated and balanced Roboflow dataset (including train/val/test splits and augmentations).
* Trained model files with evaluation metrics ($mAP@0.5$, Precision, Recall).
* Python integration script/documentation for testing on sample video streams.
**Required Qualifications:**
* Proven experience with Roboflow, YOLO (v8/v11/NAS), OpenCV, PyTorch, and TensorFlow.
* Solid track record in fine-grained object detection and handling small-object vision tasks.
* Prior experience in safety/PPE compliance projects is a strong plus.
---
**Project Description:**
We are developing an AI-driven restaurant and kitchen safety monitoring system. We need an experienced Computer Vision & Deep Learning Engineer to build, annotate, and fine-tune a high-accuracy object detection/segmentation model (using Roboflow and YOLO models like YOLOv8/v11) to accurately detect specific hygiene violations in real-time CCTV feeds.
**Key Challenges & Core Requirements:**
1. **Glove Compliance:**
* Detect bare hands vs. gloved hands (handling fine-grained vision tasks on food prep surfaces).
* Accurately flag "No Gloves" violations when staff are touching/preparing food.
2. **Proper Mask Wearing:**
* Detect proper mask usage vs. improper usage (mask pulled down to the chin / nose exposed) vs. no mask.
3. **Hairnet / Head Cover Detection:**
* High-precision detection of kitchen hairnets/caps on cooks, distinguishing between bare head, improper wear, and full coverage under varying lighting and camera angles.
4 Mobile Phone Usage:
Detect employees holding or using smartphones while working/preparing food.
Differentiate between a phone held to the ear, phone held in hand near food prep stations, and hand-only gestures.
**Scope of Work:**
* Review, clean, and augment our existing Roboflow dataset or assist in sourcing/annotating edge-case images.
* Define a robust labeling framework (e.g., multi-class classification or 2-stage object detection: Person $\rightarrow$ Face/Hands $\rightarrow$ Compliance state).
* Train and fine-tune a SOTA detection model (YOLOv8/v11, Roboflow Workflows, or custom PyTorch pipeline).
* Optimize for high Precision and Recall on edge cases (low light, overhead angles, fast hand movement).
* Export model weights (ONNX, PyTorch, or Roboflow Hosted API) for seamless integration into our existing codebase.
**Deliverables:**
* Fully annotated and balanced Roboflow dataset (including train/val/test splits and augmentations).
* Trained model files with evaluation metrics ($mAP@0.5$, Precision, Recall).
* Python integration script/documentation for testing on sample video streams.
**Required Qualifications:**
* Proven experience with Roboflow, YOLO (v8/v11/NAS), OpenCV, PyTorch, and TensorFlow.
* Solid track record in fine-grained object detection and handling small-object vision tasks.
* Prior experience in safety/PPE compliance projects is a strong plus.
---
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.