Spiking-YOLO: Spiking Neural Network for Energy-Efficient Object Detection

Authors: Seijoon Kim, Seongsik Park, Byunggook Na, Sungroh Yoon11270-11277

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Our experiments show that Spiking-YOLO achieves remarkable results that are comparable (up to 98%) to those of Tiny YOLO on non-trivial datasets, PASCAL VOC and MS COCO.
Researcher Affiliation Academia Seijoon Kim, Seongsik Park, Byunggook Na, Sungroh Yoon Department of Electrical and Computer Engineering, ASRI, INMC, and Institute of Engineering Research Seoul National University, Seoul 08826, Korea sryoon@snu.ac.kr
Pseudocode Yes Algorithm 1: Channel-wise normalization
Open Source Code No The paper does not provide any explicit statements about releasing source code or include links to a code repository.
Open Datasets Yes Our experiments show that Spiking-YOLO achieves remarkable results that are comparable (up to 98%) to those of Tiny YOLO on non-trivial datasets, PASCAL VOC and MS COCO.
Dataset Splits No No explicit percentages or sample counts for training/validation/test splits are provided. The paper mentions training on 'PASCAL VOC and MS COCO' and states 'maximum activation is calculated from the training dataset', but does not specify how the datasets were split into train, validation, and test sets to ensure reproducibility.
Hardware Specification Yes Our simulation is based on the Tensor Flow Eager and we conducted all experiments on NVIDIA Tesla V100 GPUs.
Software Dependencies No Our simulation is based on the Tensor Flow Eager. The paper does not provide specific version numbers for TensorFlow Eager or any other software libraries.
Experiment Setup Yes For further analysis, we performed additional experiments on two different output decoding schemes: one based on accumulated Vmem, and another based on spike count.