This is a re-trained version of Faster R-CNN object detection network trained with COCO* training dataset. The actual implementation is based on Detectron, with additional network weight pruning applied to sparsify convolution layers (60% of network parameters are set to zeros).
The model input is a blob that consists of a single image of "1x3x800x1280" in BGR order. The pixel values are integers in the [0, 255] range.
Metric | Value |
---|---|
Mean Average Precision (mAP) | 38.74%** |
Flops | 364.21Bn |
MParams | 52.79 |
Source framework | TensorFlow* |
Average Precision metric described in: "COCO: Common Objects in Context". The primary challenge metric is used. Tested on COCO validation dataset.
image_id
, class_id
, confidence
, x0
, y0
, x1
, y1
], respectively.image_id
- image ID in the batchclass_id
- predicted class IDconfidence
- [0, 1] detection score, the higher the value, the more confident the deteciton is onx0
, y0
) - normalized coordinates of the top left bounding box corner, in range of [0, 1]x1
, y1
) - normalized coordinates of the bootm right bounding box corner, in range of [0, 1].[*] Other names and brands may be claimed as the property of others.
[**] May be different from the original implementation due to different input configurations.