Image retrieval model based on MobileNetV2 architecture as a backbone.
| Metric | Value |
|---|---|
| Top1 accuracy | 0.834 |
| GFlops | 0.613 |
| MParams | 2.535 |
| Source framework | TensorFlow* |
Image, name: Placeholder, shape: 1, 224, 224, 3 in the format B, H, W, C, where:
B- batch sizeH- image heightW- image widthC- number of channels
Expected color order: BGR.
Tensor with name model/tf_op_layer_l2_normalize/l2_normalize and the shape 1, 256 — image embedding vector.
The model can be used in the following demos provided by the Open Model Zoo to show its capabilities:
[*] Other names and brands may be claimed as the property of others.