Jacob Solawetz

10 Jul 2020 • 5 min read

SUMMARY This guide walks through exporting an annotated dataset from Supervisely and converting its custom JSON format to YOLO Darknet format for use outside the Supervisely platform. The process involves downloading a tar archive of images and per-image annotation files, parsing the bounding box coordinates from Supervisely's

1 Jul 2020 • 6 min read

Learn how to train a YOLOv4-tiny model on a custom dataset.

18 Jun 2020 • 4 min read

SUMMARY Google's Open Images dataset contains millions of labeled images across hundreds of object classes, all under a CC BY 4.0 license, but downloading the full dataset requires over 500 GB of storage. This post shows how to use the OIDv4_ToolKit to query specific classes and

12 Jun 2020 • 16 min read

SUMMARY When Roboflow published an early YOLOv5 benchmark in June 2020, the machine learning community raised two substantive objections: whether naming Glenn Jocher's repository YOLOv5 was appropriate given no paper was published, and whether the benchmark methodology was reproducible and fair. This post documents the full history of

10 Jun 2020 • 4 min read

SUMMARY Released in June 2020, YOLOv5 is a PyTorch-based object detection model from Ultralytics that differs from YOLOv4 in several practical ways: it uses a lighter Ultralytics PyTorch framework instead of Darknet, its smallest variant (YOLOv5s) weighs 27 MB compared to 244 MB for YOLOv4, and it offers faster

2 Jun 2020 • 7 min read

SUMMARY Data augmentation artificially expands training sets by applying flips, crops, rotations, noise, and occlusion techniques, and it has a measurable impact on object detection accuracy even when collecting more labeled images is not practical. RF-DETR relies heavily on augmentation to achieve its COCO benchmark results, and the same

4 May 2020 • 6 min read

SUMMARY Self-driving systems fail at edge cases like a white truck against a bright sky because object detection models learn only from the labeled examples they are trained on, making data collection and annotation the most critical and expensive part of building autonomous perception. This post walks through how

1 May 2020 • 6 min read

SUMMARY This post compares YOLOv3 and EfficientDet across four practical dimensions: training time, model size, inference speed, and accuracy on two custom datasets (chess piece detection and blood cell detection). Both models were implemented in PyTorch and evaluated using the same Colab environment to reflect realistic deployment conditions rather than

22 Apr 2020 • 7 min read

SUMMARY EfficientDet, published by the Google Brain team, builds on EfficientNet by combining compound model scaling with a bi-directional feature pyramid network (BiFPN) for multi-scale feature fusion. The result is an object detection architecture that outperforms similarly sized models like YOLOv3, Faster R-CNN, and MobileNet on COCO