Learning Coarse-to-Fine Structured Feature Embedding for Vehicle Re-Identification

Authors: Haiyun Guo, Chaoyang Zhao, Zhiwei Liu, Jinqiao Wang, Hanqing Lu

AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on Vehicle-1M and Vehicle ID demonstrate the superiority of our proposed approach.
Researcher Affiliation Academia National Laboratory of Pattern Recognition, Institute of Automation University of Chinese Academy of Sciences, Beijing, China, 100190 {haiyun.guo, chaoyang.zhao, zhiwei.liu, jqwang, luhq}@nlpr.ia.ac.cn
Pseudocode No The paper does not contain any pseudocode or algorithm blocks.
Open Source Code No The paper does not provide an explicit statement or link for the open-source code of the described methodology. The provided URL is for the dataset, not the code.
Open Datasets Yes The paper introduces Vehicle-1M, stating ‘Available at http://www.nlpr.ia.ac.cn/iva/homepage/jqwang/Vehicle1M.htm.’ It also utilizes the Vehicle ID dataset, previously released and commonly used for vehicle re-ID.
Dataset Splits No The paper mentions using a ‘validation set’ during parameter tuning (‘From the output loss value on the validation set we found the classification loss value...’), but it does not provide specific details on its size, proportion, or how it was split from the main dataset for reproduction.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments, such as GPU models, CPU types, or memory.
Software Dependencies No The paper does not provide specific software dependency details with version numbers (e.g., Python version, PyTorch/TensorFlow versions, CUDA version).
Experiment Setup Yes The paper specifies several hyperparameters including distance margins (Mc = Mf = 0.2), K1 = 10, K2 = 3, mini-batch size = 150, and weighting values for loss terms (α = 100, β = 1000, γ = 10).