Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer
Authors: Tao Wang, Hong Liu, Pinhao Song, Tianyu Guo, Wei Shi2540-2549
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments over five challenging datasets for two tasks (occluded and holistic Re-ID) demonstrate that our proposed PFD is superior promising, which performs favorably against state-of-the-art methods. |
| Researcher Affiliation | Academia | Key Laboratory of Machine Perception Peking University, Shenzhen Graduate School {taowang, levigty}@stu.pku.edu.cn, {hongliu, pinhaosong, pkusw}@pku.edu.cn |
| Pseudocode | No | The paper describes the proposed method in detail using equations and textual descriptions, but it does not include a formal pseudocode or algorithm block. |
| Open Source Code | Yes | Code is available at https://github.com/Wang Tao As/PFD Net |
| Open Datasets | Yes | Occluded-Duke (Miao et al. 2019) consists of 15,618 training images, 2,210 occluded query images and 17,661 gallery images. It is a subdataset of Duke MTMC-re ID (Zheng, Zheng, and Yang 2017)... Market-1501 (Zheng et al. 2015)... Duke MTMC-re ID (Zheng, Zheng, and Yang 2017)... MSMT17 (Wei et al. 2018)... |
| Dataset Splits | Yes | Occluded-Duke (Miao et al. 2019) consists of 15,618 training images, 2,210 occluded query images and 17,661 gallery images... Market-1501 (Zheng et al. 2015) contains 1,501 identities observed from 6 camera viewpoints, 12,936 training images of 751 identities, 19,732 gallery images, and 2,228 queries. |
| Hardware Specification | No | The paper describes experimental setup and parameters but does not provide specific hardware details such as GPU models, CPU types, or memory. |
| Software Dependencies | No | The paper mentions software components like 'transformer decoder' and 'HRNet' but does not provide specific version numbers for them or any other libraries. |
| Experiment Setup | Yes | Both training and testing images are resized to 256 128. The training images are augmented with random horizontal flipping, padding, random cropping and random erasing (Zhong et al. 2020). The initial weights of encoder are pretrained on Image Net-21K and then finetuned on Image Net1K... The number of decoder layer is set to 2 on Occluded Duke and 6 on the other datasets. The hidden dimension D is set to 768. The batch size is set to 64 with 4 images per ID. The learing rate is initialized at 0.008 with cosine learning rate decay. The threshold γ is set to 0.2. |