mirrors
/
fast-reid
mirror of https://github.com/JDAI-CV/fast-reid.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

chore: update model zoo

pull/45/head
liaoxingyu 2020-04-29 18:15:27 +08:00
parent 6cee977748
commit fdaf82cd62
1 changed files with 71 additions and 9 deletions

80
MODEL_ZOO.md
View File

@ -5,27 +5,89 @@
### Market1501 Baselines
**BoT**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| BoT(R50) | ImageNet | 93.6% | 85.1% | 58.1% |
| BoT(R50-ibn) | ImageNet | 94.8% | 87.3% | 63.5% |
| BoT(S50) | ImageNet | - | - | - |
| BoT(R101-ibn) | ImageNet| - | - |
**AGW**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| BagTricks(R50) | ImageNet | 93.6% | 85.1% | 58.1% |
| BagTricks(R50-ibn) | ImageNet | 94.8% | 87.3% | 63.5% |
| AGW(R50) | ImageNet | 94.9% | 87.4% | 63.1% |
| stronger baseline(R50-ibn) | ImageNet | 95.5% | 88.4% | 65.8% |
| AGW(R50-ibn) | ImageNet | - | - | - |
| AGW(R101-ibn) | ImageNet | - | - | - |
**SBS**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| SBS(R50) | ImageNet | - | - | - |
| SBS(R50-ibn) | ImageNet | 95.5% | 88.4% | 65.8% |
| SBS(S50) | ImageNet | - | - | - |
| SBS(R101-ibn) | ImageNet | - | - | - |
### DukeMTMC Baseline
**BoT**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| BoT(R50) | ImageNet | 86.1% | 75.9% | 38.7% |
| BoT(R50-ibn) | ImageNet | 89.0% | 78.8% | 43.6% |
| BoT(S50) | ImageNet | - | - | - |
| BoT(R101-ibn) | ImageNet| - | - |
**AGW**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| BagTricks(R50) | ImageNet | 86.1% | 75.9% | 38.7% |
| BagTricks(R50-ibn) | ImageNet | 89.0% | 78.8% | 43.6% |
| AGW(R50) | ImageNet | 88.9% | 79.1% | 43.2% |
| stronger baseline(R50-ibn) | ImageNet | 91.3% | 81.6% | 47.6% |
| AGW(R50-ibn) | ImageNet | - | - | - |
| AGW(R101-ibn) | ImageNet | - | - | - |
**SBS**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| SBS(R50) | ImageNet | - | - | - |
| SBS(R50-ibn) | ImageNet | 91.3% | 81.6% | 47.6% |
| SBS(S50) | ImageNet | - | - | - |
| SBS(R101-ibn) | ImageNet | - | - | - |
### MSMT17 Baseline
**BoT**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| BoT(R50) | ImageNet | 70.4% | 47.5% | 9.6% |
| BoT(R50-ibn) | ImageNet | 76.9% | 55.0% | 13.5% |
| BoT(S50) | ImageNet | - | - | - |
| BoT(R101-ibn) | ImageNet| - | - |
**AGW**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| BagTricks(R50) | ImageNet | 70.4% | 47.5% | 9.6% |
| BagTricks(R50-ibn) | ImageNet | 76.9% | 55.0% | 13.5% |
| AGW(R50) | ImageNet | 75.6% | 52.6% | 11.9% |
| stronger baseline(R50-ibn) | ImageNet | 84.2% | 61.5% | 15.7% |
| AGW(R50-ibn) | ImageNet | - | - | - |
| AGW(R101-ibn) | ImageNet | - | - | - |
**SBS**:
| Method | Pretrained | Rank@1 | mAP | mINP |
| :---: | :---: | :---: |:---: | :---: |
| SBS(R50) | ImageNet | - | - | - |
| SBS(R50-ibn) | ImageNet | 84.2% | 61.5% | 15.7% |
| SBS(S50) | ImageNet | - | - | - |
| SBS(R101-ibn) | ImageNet | - | - | - |