Enhancing Lightweight Neural Networks for Small Object Detection in IoT Applications
This addresses the challenge of detecting small objects in embedded IoT devices, which is crucial for applications like remote monitoring, though it appears incremental as it builds on existing detectors.
The paper tackled the problem of small object detection for IoT applications by proposing an adaptive tiling method that works with existing object detectors like FOMO, achieving up to 225% improvement in F1-score and up to 76% reduction in average object count error.
Advances in lightweight neural networks have revolutionized computer vision in a broad range of IoT applications, encompassing remote monitoring and process automation. However, the detection of small objects, which is crucial for many of these applications, remains an underexplored area in current computer vision research, particularly for embedded devices. To address this gap, the paper proposes a novel adaptive tiling method that can be used on top of any existing object detector including the popular FOMO network for object detection on microcontrollers. Our experimental results show that the proposed tiling method can boost the F1-score by up to 225% while reducing the average object count error by up to 76%. Furthermore, the findings of this work suggest that using a soft F1 loss over the popular binary cross-entropy loss can significantly reduce the negative impact of imbalanced data. Finally, we validate our approach by conducting experiments on the Sony Spresense microcontroller, showcasing the proposed method's ability to strike a balance between detection performance, low latency, and minimal memory consumption.