Research map / Detection and segmentation / CNN features
DAC-SDC Low Power Object Detection Challenge for UAV Applications
Xiaowei Xu, Xinyi Zhang, Bei Yu, Xiaobo Sharon Hu, Christopher Rowen, Jingtong Hu, Yiyu Shi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use binarized layers and dynamic precision (SystemsETHZ vs TGIIF) | FPGA detector · DAC-SDC FPGA | IoU and FPS: IoU 0.6238 -> 0.4919; FPS 11.96 -> 25.97 | Mixed |
| prune SSD and quantize to 8-bit fixed point | TGIIF SSD · DAC-SDC FPGA | parameters and operations: 14.2x less parameters and 10x less operations | Improved |
| use depthwise separable convolution IP reuse (iSmart2) | FPGA detector · DAC-SDC FPGA | resource utilization: LUT 63%, Flip-Flop 22% | Improved |
| use 1-2 bit parameters instead of 8-16 bit | FPGA detector · DAC-SDC FPGA | DSP utilization: 100% -> 12% | Improved |
| use FPGA instead of GPU platform | low power detection · DAC-SDC | energy consumption: 1/3-1/2 energy consumption | Improved |
| higher image brightness | GPU entries · DAC-SDC | accuracy | Improved |
| larger amount of information in image | GPU and FPGA entries · DAC-SDC | accuracy | Improved |
| object size ratio 1% instead of 2% | GPU entries · DAC-SDC | accuracy | Hurt |
| object size above 6% ratio | GPU entries · DAC-SDC | accuracy | Hurt |
| high IoU but FPS below 20 (Talos-G) | GPU entries · DAC-SDC | ranking: ranks 14th | Hurt |
| high FPS but low IoU (OSSDC) | GPU entries · DAC-SDC | ranking: ranks 15th | Hurt |
| use half precision float and TensorRT | ICT-CAS tiny YOLO · DAC-SDC GPU | speed and power | Improved |
Closest papers by meaning
- RemDet: Rethinking Efficient Model Design for UAV Object Detection (AAAI 2025 · similarity 0.78)
- Multiview Aerial Visual RECognition (MAVREC): Can Multi-view Improve Aerial Visual Perception? (CVPR 2024 · similarity 0.77)
- Mamba YOLO: A Simple Baseline for Object Detection with State Space Model (AAAI 2025 · similarity 0.76)
- Towards Unsupervised Model Selection for Domain Adaptive Object Detection (NeurIPS 2024 · similarity 0.76)
- DETRs Beat YOLOs on Real-time Object Detection (CVPR 2024 · similarity 0.75)
- Self-Prompting Analogical Reasoning for UAV Object Detection (AAAI 2025 · similarity 0.75)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
