Science Daily May 3, 2023
Existing image-free methods still cannot simultaneously obtain the category, location, and size information of all objects. Researchers in China have reported a novel image-free single-pixel object detection (SPOD) technique which enables efficient and robust multi-object detection directly from a small number of measurements, eliminating the requirement for complicated image reconstruction. The small-size optimized pattern sampling method achieved higher image-free sensing accuracy with fewer pattern parameters than convolutional neural network (CNN). Instead of simply stacking CNN layers, they designed the SPOD network based on the transformer architecture. It can better model global features and reinforce the network’s attention to the targets in the scene, thus improving the object detection performance. According to the researchers they demonstrated the effectiveness of SPOD which achieves a detection accuracy of 82.41% mAP at a sampling rate of 5% with a refresh rate of 63 f.p.s… read more. TECHNICAL ARTICLEÂ
The researchers showed that SPOD achieves a detection accuracy of 82.41% mAP… Credit: Lintao Peng, Beijing Institute of Technology