DeNA open-sources PyTorch implementation of YOLOv3 object detector
DeNA Co., Ltd. is proud to announce open-sourcing of PyTorch_YOLOv3, a re-implementation of the object detector YOLOv3 in PyTorch. Our implementation reproduces the detection performance of the original YOLOv3 written in C.
YOLOv3 is one of the state-of-the-art object detectors that is capable of real-time and accurate detection of multiple objects (e. g. people, vehicles and animals) in images. Hiroto Honda, an AI research engineer at DeNA, has re-implemented the detector in PyTorch, which is rapidly gaining popularity among deep learning developers, and has successfully reproduced detection accuracy in both training and test phases.
We are excited to contribute to the open-source community and hope that computer vision researchers and developers can benefit from the project.