Learning Resolution-Invariant Deep Representations for Person Re-Identification
Authors: Yun-Chun Chen, Yu-Jhe Li, Xiaofei Du, Yu-Chiang Frank Wang8215-8222
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our experiments confirm that the use of our model can recognize low-resolution query images, even if the resolution is not seen during training. Moreover, the extension of our model for semi-supervised re-ID further confirms the scalability of our proposed method for real-world scenarios and applications.Extensive experiments are performed to verify the effectiveness of our model, and confirm its use for re-ID in semi-supervised settings.We perform evaluations on three benchmark datasets, including two synthetic and one real-world person re-ID datasets.Table 1 lists the quantitative results on the three datasets. |
| Researcher Affiliation | Collaboration | Yun-Chun Chen,1 Yu-Jhe Li,1,2 Xiaofei Du,3 Yu-Chiang Frank Wang1,2 1Department of Electrical Engineering, National Taiwan University 2MOST Joint Research Center for AI Technology and All Vista Healthcare 3Umbo Computer Vision Email: {b03901148, r06942074, ycwang}@ntu.edu.tw, xiaofei.du@umbocv.com |
| Pseudocode | No | The paper does not contain any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not include any statement about releasing open-source code or a link to a code repository for the methodology described. |
| Open Datasets | Yes | We perform evaluations on three benchmark datasets, including two synthetic and one real-world person re-ID datasets. MLR-CUHK03. The MLR-CUHK03 dataset is a synthetic dataset built from CUHK03 (Li et al. 2014) which consists of 5 different camera views with more than 14, 000 images of 1, 467 person identities. MLR-VIPe R. The MLR-VIPe R dataset is a synthetic dataset built from VIPe R (Gray and Tao 2008) which contains 632 person-image pairs captured by two cameras. CAVIAR. The more challenging CAVIAR dataset (Cheng et al. 2011) is a genuine LR person re-ID dataset which contains 1, 220 images of 72 person identities captured from two camera views. |
| Dataset Splits | No | The paper describes how the datasets are split into training and test sets but does not explicitly mention a separate validation set split or its proportions. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used to run the experiments, such as GPU models, CPU types, or memory specifications. |
| Software Dependencies | No | The paper mentions adopting "Res Net50 (He et al. 2016)" but does not provide specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions, or other library versions). |
| Experiment Setup | No | The paper mentions using ResNet50 as the feature extractor and various loss functions. It also mentions selecting feature levels {4, 5} for the discriminator. However, it does not provide specific training hyperparameters such as learning rate, batch size, optimizer details, or the number of epochs, which are crucial for reproducing the experimental setup. |