Label-Attended Hashing for Multi-Label Image Retrieval

Authors: Yanzhao Xie, Yu Liu, Yangtao Wang, Lianli Gao, Peng Wang, Ke Zhou

IJCAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on public multi-label datasets demonstrate that (1) LAH can achieve the state-of-the-art retrieval results and (2) the usage of co-occurrence relationship and MFB not only promotes the precision of hash codes but also accelerates the hash learning.
Researcher Affiliation Academia 1Huazhong University of Science and Technology 2The University of Electronic Science and Technology of China
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks (clearly labeled algorithm sections or code-like formatted procedures).
Open Source Code Yes Git Hub address: https: //github.com/IDSM-AI/LAH
Open Datasets Yes VOC2007. [Everingham et al., 2010] consists of 9,963 multi-label images and 20 object classes. MS-COCO. [Lin et al., 2014] is a popular multiple object dataset for image recognition, segmentation and captioning, which contains 118,287 training images, 40,504 validation images and 40,775 test images... FLICKR25K. [Huiskes and Lew, 2008] is a collection of 25,000 multi-label images belonging to 24 unique provided labels...
Dataset Splits Yes MS-COCO... contains 118,287 training images, 40,504 validation images and 40,775 test images... In the part of fch, we set the model parameters (γ = 1 and λ = 0.55) of LAH by cross-validation.
Hardware Specification No The paper mentions using PyTorch for implementation but does not specify any particular hardware details such as CPU/GPU models or memory.
Software Dependencies No The processing is implemented using Py Torch1. For network optimization, Stochastic Gradient Descent (SGD) [Amari, 1993] is used as the optimizer. Specific version numbers for PyTorch or other libraries are not provided.
Experiment Setup Yes For label co-occurrence embedding learning, our LAH consists of two GCN layers with output dimensionality of 1024 and 2048... we set τ = 0.4 and q = 0.2. we adopt Res Net-101 pre-trained on Image Net... mini-batch size is fixed as 256 and the raw images (input) are random resized into 448 448 using random horizontal flips. In the part of MFB, ... we set k = 350 for all datasets. For fair comparisons with other algorithms, we set G = 350. ...we set the model parameters (γ = 1 and λ = 0.55)... SGD... with 0.9 momentum and 10 4 weight decay. Note that all the results are obtained within 20 epochs.