Google Summer of Code: Week 4

This week was an eventful one. I started the week with TF Object Detection model development. I trained two models one was Faster RCNN and another one was MobileNet SSD. I plan to train YOLO models too and convert them into ONNX format for JS inference.

For VR the bounding boxes posed a lot of problems . For rendering the bounding boxes I used HTML5 Canvas and WebGL for designing and rendering the bounding box. The Dataset for Segmentation has only one class so class labels are not rendered on top of the bounding box. The model inference is still slow and in some cases fails to render the bounding boxes.

POC for Object detection is not working properly at this juncture. It still requires few more corrections for improving the model performance.

That's all for today!!!

Hope you had a great week

Vale

Show Comments