YOLOv11 Architecture

Previously we mentioned its younger brother YOLOv10. Today we continue with YOLOv11, the newest in the series. YOLO is an almost unrivaled algorithm that produces very successful results in the field of object detection. This algorithm series, which continues to be developed by Ultralytics after YOLOv5, continues to produce better performance with each new model.

For readers who can’t see the full story, click here.

Made by GPT

YOLOv11 is the latest YOLO model developed by Ultralytics. This model continues to balance accuracy and efficiency while performing real-time object detection. Building on previous YOLO versions, YOLO11 offers significant improvements in architecture and training. The most important architectural change that improves performance while maintaining speed is the addition of the C3K2 block, the SPFF module and the C2PSA block.

C3K2 block: This is an enhancement of the CSP (Cross Stage Partial) block introduced in previous versions. This module optimizes the extraction of more complex features using different kernel sizes (e.g. 3x3 or 5x5) and channel separation strategies.

SPFF (Spatial Pyramid Pooling Fusion) module: It is an optimized version of the SPP (Spatial Pyramid Pooling) module used in YOLO versions. This module allows the model to perform better by capturing the properties of objects at different scales.