Retrieving Object-Level Features From YOLO
Extract the object-level features from YOLO for downstream tasks without extra overhead.
I develop computer vision solutions for video analytics at the edge.
Extract the object-level features from YOLO for downstream tasks without extra overhead. Implement class balancing in Ultralytics using a weighted dataloader and improve the performance of minority class. Get over 10% more mAP in small object detection by exploiting YOLOv8 pose models while training. Add additional classes to pre-trained YOLOv8 model without affecting the confidences on the existing classes. Some useful tricks and hacks I learnt while using MMDetection for my research work.Recent Posts
Retrieving Object-Level Features From YOLO
Balance Classes During YOLO Training Using a Weighted Dataloader
A Simple Trick To Increase YOLOv8's Accuracy On Small Objects With No Overhead
Extending YOLOv8 COCO Model With New Classes Without Affecting Old Weights
MMDetection 3.x Hacks and Tricks
My Projects
View all »