cuulee Goto Github PK

followers: 8.0 following: 1.0 repos: 7.0K gists: 70.0

Type: User

cuulee's Projects

tensorflow-image-classification

CNN for multi-class image recognition in tensorflow

tensorflow-lite-object-detection-on-android-and-raspberry-pi

A tutorial showing how to train, convert, and run TensorFlow Lite object detection models on Android devices, the Raspberry Pi, and more!

tensorflow-object-detection-api-google_colab

Mask Detection as my contribution for Covid-19.

tensorflow-object-detection-on-the-raspberry-pi

A tutorial showing how to set up TensorFlow's Object Detection API on the Raspberry Pi

tensorflow_psenet

This is a tensorflow re-implementation of PSENet: Shape Robust Text Detection with Progressive Scale Expansion Network.My blog:

Sliding-window object detectors that generate boundingbox object predictions over a dense, regular grid have advanced rapidly and proven popular. In contrast, modern instance segmentation approaches are dominated by methods that first detect object bounding boxes, and then crop and segment these regions, as popularized by Mask R-CNN. In this work, we investigate the paradigm of dense slidingwindow instance segmentation, which is surprisingly underexplored. Our core observation is that this task is fundamentally different than other dense prediction tasks such as semantic segmentation or bounding-box object detection, as the output at every spatial location is itself a geometric structure with its own spatial dimensions. To formalize this, we treat dense instance segmentation as a prediction task over 4D tensors and present a general framework called TensorMask that explicitly captures this geometry and enables novel operators on 4D tensors. We demonstrate that the tensor view leads to large gains over baselines that ignore this structure, and leads to results comparable to Mask R-CNN. These promising results suggest that TensorMask can serve as a foundation for novel advances in dense mask prediction and a more complete understanding of the task. Code will be made available.