|
| 1 | +# Fast Image Retrieval |
| 2 | + |
| 3 | +## Introduction |
| 4 | +Fast Image Retrieval is an open source image retrieval framework release by Center of Image and Signal Processing Lab(CISiP Lab), Universiti Malaya. This framework implemented most major binary hashing methods, together with major backbone networks and major datasets. |
| 5 | + |
| 6 | +### Major features |
| 7 | +- **One for All** |
| 8 | + |
| 9 | + We unified various binary hashing methods, different backbone, and multiple datasets under a same single framework, easing the benchmark and research. It supports popular binary hashing methods, e.g. HashNet, GreedyHash, DPN, OrthoHash, etc. |
| 10 | +- **Modularity** |
| 11 | + |
| 12 | + We break the framework into parts so that one can easily implement their own method by joining up the components. |
| 13 | + |
| 14 | +## License |
| 15 | +[comment]: <> (Add license) |
| 16 | +This project is released under [BSD-3 license](). |
| 17 | +## Changelog |
| 18 | + |
| 19 | +[comment]: <> (Change to docs link) |
| 20 | +Please refer to [Changelog](?) for more detail. |
| 21 | +## Implemented method/backbone/datasets |
| 22 | +[//]: <> (This is copied from docs/implementation.md) |
| 23 | +### Backbone |
| 24 | +1. Alexnet |
| 25 | +2. VGG{16} |
| 26 | +3. ResNet{18,34,50,101,152} |
| 27 | + |
| 28 | +### Loss (Method) |
| 29 | +#### Supervised |
| 30 | +|Method|Config Template|Loss Name| |
| 31 | +|---|---|---| |
| 32 | +|ADSH|adsh.yaml|adsh| |
| 33 | +|BiHalf|bihalf-supervised.yaml|bihalf-supervised| |
| 34 | +|Cross Entropy|ce.yaml|ce| |
| 35 | +|CSQ|csq.yaml|csq| |
| 36 | +|DBDH|dbdh.yaml|dbdh| |
| 37 | +|DFH|dfh.yaml|dfh| |
| 38 | +|DPN|dpn.yaml|dpn| |
| 39 | +|DPSH|dpsh.yaml|dpsh| |
| 40 | +|DTSH|dtsh.yaml|dtsh| |
| 41 | +|GreedyHash|greedyhash.yaml|greedyhash| |
| 42 | +|HashNet|hashnet.yaml|hashnet| |
| 43 | +|JMLH|jmlh.yaml|jmlh| |
| 44 | +|MIHash|mihash.yaml|mihash| |
| 45 | +|OrthoCos(OrthoHash)|orthocos.yaml|orthocos| |
| 46 | +|OrthoArc(OrthoHash)|orthoarc.yaml|orthoarc| |
| 47 | +|SDH|sdh.yaml|sdh| |
| 48 | +|SDH-C|sdhc.yaml|sdhc| |
| 49 | +#### Unsupervised |
| 50 | +|Method|Config Template|Loss Name| |
| 51 | +|---|---|---| |
| 52 | +|BiHalf|bihalf.yaml|bihalf| |
| 53 | +|CIBHash|cibhash.yaml|cibhash| |
| 54 | +|GreedyHash|greedyhash-unsupervised.yaml|greedyhash-unsupervised| |
| 55 | +|SSDH|ssdh.yaml|ssdh| |
| 56 | +|TBH|tbh.yaml|tbh| |
| 57 | +#### Shallow (Non-Deep learning methods) |
| 58 | +|Method|Config Template|Loss Name| |
| 59 | +|---|---|---| |
| 60 | +|ITQ|itq.yaml|itq| |
| 61 | +|LsH|lsh.yaml|lsh| |
| 62 | +|PCAHash|pca.yaml|pca| |
| 63 | +|SH|sh.yaml|sh| |
| 64 | + |
| 65 | + |
| 66 | + |
| 67 | +### Datasets |
| 68 | +|Dataset|Name in framework| |
| 69 | +|---|---| |
| 70 | +|ImageNet100|imagenet| |
| 71 | +|NUS-WIDE|nuswide| |
| 72 | +|MS-COCO|coco| |
| 73 | +|MIRFLICKR/Flickr25k|mirflickr| |
| 74 | +|Stanford Online Product|sop| |
| 75 | +|Cars dataset|cars| |
| 76 | +|CIFAR10|cifar10| |
| 77 | + |
| 78 | +## Installation |
| 79 | +[comment]: <> (Change to docs link) |
| 80 | +Please head up to [get_started.md](docs/get_started.md) for guides on setup conda environment and installation. |
| 81 | +## Tutorials |
| 82 | +[comment]: <> (Change to docs link) |
| 83 | +Please head up to [tutorials.md](docs/tutorials.md) for guidances. |
| 84 | + |
| 85 | +[comment]: <> (## Citation) |
| 86 | + |
| 87 | +[comment]: <> (If you find this framework useful in your research, please consider cite this project.) |
| 88 | + |
| 89 | +[comment]: <> (```) |
| 90 | + |
| 91 | +[comment]: <> (@misc{) |
| 92 | + |
| 93 | + |
| 94 | +[comment]: <> (```) |
| 95 | +## Contributing |
| 96 | +We welcome the contributions to improve this project. Please file your suggestions/issues by creating new issues or send us a pull request for your new changes/improvement/features/fixes. |
0 commit comments