♻️ Introducing CircularNet, a set of data efficient models created to support the way we manage and recycle materials across the waste management ecosystem.
Learn how CircularNet uses TensorFlow's Mask R-CNN algorithm to change the future of recycling ➡️ goo.gle/3shYqgA
Once data is collected at Material Recovery Facilities,
✅ Annotation files get converted into COCO JSON
✅ Incorrect labels, errors and corrupt images are removed to ensure smooth training.
✅ Final file gets converted to the TF record format
The Mask RCNN is then trained using the Model Garden Repository.
✅ Hyper parameter optimization is performed by changing image size, batch
🏁 Final checkpoints are achieved once the model is trained and converted into saved model and TFLite formats.
This allows the model to support server side and edge side deployments on Google Cloud.
🔎 Through visualization, the model is able to identify:
🟢 Material type
🟢 Product form of material
🟢 Types of plastics, and more!
If you want to learn more about how CircularNet utilizes #ML to reduce waste, click here → goo.gle/3shYqgA
• • •
Missing some Tweet in this thread? You can try to
force a refresh
💡The first demo, Mind Controller, combines hardware and software using an image classifier to control an IoT device.
🏎️ The second demo, Eye Driver, combines two very different machine learning models trained in two completely different tasks to achieve controlling a video game with just your eyes!