YOLIC: An Efficient Method for Object Localization and Classification on Edge Devices

TL;DR

YOLIC is a novel, efficient object localization and classification method optimized for edge devices, combining segmentation and detection techniques to achieve high speed and accuracy without bounding box regression.

Contribution

YOLIC introduces a new cell-based approach that simplifies object detection on edge devices, eliminating the need for bounding box regression and enabling multi-label classification for overlapping objects.

Findings

Achieves detection performance comparable to YOLO algorithms.

Surpasses 30fps on Raspberry Pi 4B CPU.

Reduces computational load while maintaining accuracy.

Abstract

In the realm of Tiny AI, we introduce ``You Only Look at Interested Cells" (YOLIC), an efficient method for object localization and classification on edge devices. Through seamlessly blending the strengths of semantic segmentation and object detection, YOLIC offers superior computational efficiency and precision. By adopting Cells of Interest for classification instead of individual pixels, YOLIC encapsulates relevant information, reduces computational load, and enables rough object shape inference. Importantly, the need for bounding box regression is obviated, as YOLIC capitalizes on the predetermined cell configuration that provides information about potential object location, size, and shape. To tackle the issue of single-label classification limitations, a multi-label classification approach is applied to each cell for effectively recognizing overlapping or closely situated objects. This paper presents extensive experiments on multiple datasets to demonstrate that YOLIC achieves detection performance comparable to the state-of-the-art YOLO algorithms while surpassing in speed, exceeding 30fps on a Raspberry Pi 4B CPU. All resources related to this study, including datasets, cell designer, image annotation tool, and source code, have been made publicly available on our project website at https://kai3316.github.io/yolic.github.io