Code for Enhancing Generalization in Sketch-Based Image Retrieval through Single and Multi-Source Domain Adaptation

464 14 2430 2 8421 document 2430 Thesis code.zip application/zip c21828aba3543e837d82bbe6152f7bd2 MD5 15693747 2025-06-17 17:27:50 https://bordar.bournemouth.ac.uk/464/1/Thesis%20code.zip 464 1 1 application/zip archive en public cc_by_nd_4

Thesis code.zip

full_archive archive 11424 disk0/00/00/04/64 2025-10-29 13:36:45 2025-10-29 13:37:06 2025-10-29 13:36:45 data_collection show Huang Mengqing s5078786@bournemouth.ac.uk Code for Enhancing Generalization in Sketch-Based Image Retrieval through Single and Multi-Source Domain Adaptation media Sketch-Based Image Retrieval, Domain Adaptation, Generalization, Deep Learning, Transfer Learning This study is conducted on the CIFAR-100 dataset, with only 50 randomly selected object classes used for training. The remaining classes are reserved for zero-shot testing. Each pre-trained model is fine-tuned on CIFAR-100 to adapt itself to this task. Subsequently, various model variants are created by adding a single linear probe layer for subsequent zero-shot performance evaluation. These models are then tested, collecting metrics such as error rate and kappa. Finally, the trade-off point between these metrics is calculated and visualized in graphs, ready for analysis. 2025-06-17 published 10.18746/bmth.data.00000464 Code bordar@bournemouth.ac.uk Huang Mengqing s5078786@bournemouth.ac.uk A practical generalization metric for deep networks benchmarking pub http://dx.doi.org/10.1038/s41598-025-93005-5 DOI https://www.nature.com/articles/s41598-025-93005-5 pub Springer Nature Enhancing Generalization in Sketch-Based Image Retrieval through Single and Multi-Source Domain Adaptation thesis https://eprints.bournemouth.ac.uk/41316/ pub Bournemouth University 2024-01-18 2025-01-12