YOGA: Deep Object Detection in the Wild with Lightweight Feature Learning and Multiscale Attention

TL;DR

YOGA is a lightweight, scalable object detection model that achieves a high accuracy-to-size ratio on edge devices by using efficient feature learning and attention mechanisms.

Contribution

The paper introduces YOGA, a novel lightweight object detector with a two-phase feature learning pipeline and multiscale attention, optimized for low-end edge devices.

Findings

YOGA achieves up to 22% higher AP compared to other models.

YOGA reduces parameters and FLOPs by 23-34%.

YOGA performs well on NVIDIA Jetson Nano.

Abstract

We introduce YOGA, a deep learning based yet lightweight object detection model that can operate on low-end edge devices while still achieving competitive accuracy. The YOGA architecture consists of a two-phase feature learning pipeline with a cheap linear transformation, which learns feature maps using only half of the convolution filters required by conventional convolutional neural networks. In addition, it performs multi-scale feature fusion in its neck using an attention mechanism instead of the naive concatenation used by conventional detectors. YOGA is a flexible model that can be easily scaled up or down by several orders of magnitude to fit a broad range of hardware constraints. We evaluate YOGA on COCO-val and COCO-testdev datasets with other over 10 state-of-the-art object detectors. The results show that YOGA strikes the best trade-off between model size and accuracy (up to 22% increase of AP and 23-34% reduction of parameters and FLOPs), making it an ideal choice for deployment in the wild on low-end edge devices. This is further affirmed by our hardware implementation and evaluation on NVIDIA Jetson Nano.