Pytorch Implementation of Unifying Deep Local and Global Features for Image Search (delg-eccv20)
- DELG pipline:
Install Python dependencies:
pip install -r requirements.txt
Set PYTHONPATH:
export PYTHONPATH=`pwd`:$PYTHONPATH
Training a delg model:
python train_delg.py \
--cfg configs/metric/resnet_delg_8gpu.yaml \
OUT_DIR ./output \
PORT 12001 \
TRAIN.WEIGHTS path/to/pretrainedmodel
Resume training:
python train_delg.py \
--cfg configs/metric/resnet_delg_8gpu.yaml \
OUT_DIR ./output \
PORT 12001 \
TRAIN.AUTO_RESUME True
-r50-delg (wu46)
-r101-delg (5pdj)
pretrained weeights are available in pymetric
!!! Queries should be cropped as DOLG.
Extracting global and local feature for multi-scales
python tools/extractor.py --cfg configs/resnet_delg_8gpu.yaml
Refer extractor.sh
for using multicards
See visualize.ipynb
for verification of local features
-
Spatial Verification
Install pydegensac and see tools/rerank/spatial_verification.py
-
Examples
See (https://github.com/filipradenovic/revisitop) for details
cd tools/revisitop
python example_evaluate_with_local.py main
- on roxford5k
Backbone | Train Size | Method | mAP E | mAP M | mAP H |
---|---|---|---|---|---|
ResNet50 | 224 | Global Ranking | 77.73 | 66.06 | 38.37 |
ResNet50 | 224 | Global | 81.03 | 68.31 | 39.98 |
ResNet50 | 224 | Global + Spatial Verification | 84.81 | 71.97 | 46.63 |
ResNet50 | 512 | Global | 90.55 | 78.51 | 56.90 |
ResNet50 | 512 | Global + Spatial Verification | 90.86 | 80.08 | 58.42 |
- on rparis6k(updating)
- SOTA of R50-DELG is 78.3 mAP@M in the paper, we outperform it
- All training set version is GLDv2-clean (81313, 1580470)
- Traing size, global and local feature scales adopted are same with the paper