A General Rank Preserving Framework for Asymmetric Image Retrieval
Authors: Hui Wu, Min Wang, Wengang Zhou, Houqiang Li
ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on various large-scale datasets demonstrate the superiority of our two proposed methods. Comprehensive experiments are conducted on four popular retrieval datasets. Ablations demonstrate the effectiveness and generalizability of our framework. Our approach surpasses the existing state-of-the-art methods by a considerable margin. |
| Researcher Affiliation | Academia | Hui Wu1 Min Wang2 Wengang Zhou1,2 Houqiang Li1,2 1CAS Key Laboratory of Technology in GIPAS, University of Science and Technology of China 2Institute of Artificial Intelligence, Hefei Comprehensive National Science Center |
| Pseudocode | Yes | Algorithm 1 Pseudo-code of Rank Preserving Framework in a Py Torch-like style. |
| Open Source Code | No | The paper does not provide an explicit statement or link indicating that the source code for the described methodology is publicly available. |
| Open Datasets | Yes | Training datasets. Two datasets are used for training. One is Sf M-120k (Radenovi c et al., 2018b), of which 551 3D models are taken for training while the other 162 3D models for validation. The other is GLDv2 (Weyand et al., 2020), which consists of 1, 580, 470 images with 81, 311 classes. We randomly sample 80% images from GLDv2 for training and leave the rest 20% for validation. |
| Dataset Splits | Yes | Training datasets. Two datasets are used for training. One is Sf M-120k (Radenovi c et al., 2018b), of which 551 3D models are taken for training while the other 162 3D models for validation. The other is GLDv2 (Weyand et al., 2020), which consists of 1, 580, 470 images with 81, 311 classes. We randomly sample 80% images from GLDv2 for training and leave the rest 20% for validation. |
| Hardware Specification | Yes | The query model is trained on one NVIDIA RTX 3090 GPU. ... We train the query model on 4 NVIDIA RTX 3090 GPUs for 10 epochs with a batch size of 256. |
| Software Dependencies | No | The paper mentions 'Py Torch-like style' for its pseudocode, but does not provide specific version numbers for any software dependencies (e.g., PyTorch, CUDA, Python version). |
| Experiment Setup | Yes | Training epochs and batch size are set as 10 and 64, respectively. ... All models are optimized using Adam with an initial learning rate of 10 3 and a weight decay of 10 6. A linearly decaying scheduler is adopted to gradually decay the learning rate to 0 when the desired number of steps is reached. When query model is trained with Rank Order Preservation (Sec. 4.1), the length K of ranking list is set to 4096, and the temperature coefficient τr in ranking weight Wi is set as 0.2. As for Monotonic Similarity Preservation (Sec. 4.2), K is also set as 4096, and both τg and τq are set to 0.1. |