Pose Guided Person Image Generation
Authors: Liqian Ma, Xu Jia, Qianru Sun, Bernt Schiele, Tinne Tuytelaars, Luc Van Gool
NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results on both 128 64 re-identification images and 256 256 fashion photos show that our model generates high-quality person images with convincing details. Experiments on two dataset, a low-resolution person re-identification dataset and a high-resolution fashion photo dataset, demonstrate the effectiveness of the proposed method. |
| Researcher Affiliation | Collaboration | 1KU-Leuven/PSI, TRACE (Toyota Res in Europe) 2KU-Leuven/PSI, IMEC 3Max Planck Institute for Informatics, Saarland Informatics Campus 4ETH Zurich |
| Pseudocode | No | The paper does not contain structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement or link about open-sourcing the code for the described methodology. |
| Open Datasets | Yes | The Deep Fashion (In-shop Clothes Retrieval Benchmark) dataset [16] consists of 52,712 in-shop clothes images, and 200,000 cross-pose/scale pairs. We also experiment on a more challenging re-identification dataset Market-1501 [37] containing 32,668 images of 1,501 persons captured from six disjoint surveillance cameras. |
| Dataset Splits | Yes | In the train set, we have 146,680 pairs each of which is composed of two images of the same person but different poses. We randomly select 12,800 pairs from the test set for testing. All images have size 128 64 and are split into train/test sets of 12,936/19,732 following [37]. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'Adam [13] optimizer' and a 'state-of-the-art pose estimator [2]' but does not specify version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | On both datasets, we use the Adam [13] optimizer with β1 = 0.5 and β2 = 0.999. The initial learning rate is set to 2e-5. On Deep Fashion, we set the number of convolution blocks N = 6. Models are trained with a minibatch of size 8 for 30k and 20k iterations respectively at stage-I and stage-II. On Market-1501, we set the number of convolution blocks N = 5. Models are trained with a minibatch of size 16 for 22k and 14k iterations respectively at stage-I and stage-II. |