@@ -9,10 +9,14 @@ SOLOv2 (Segmenting Objects by Locations) is a fast instance segmentation framewo
- Performance: `Light-R50-VD-DCN-FPN` model reached 38.6 FPS on single Tesla V100, and mask ap on the COCO-val dataset reached 38.8, which increased inference speed by 24%, mAP increased by 2.4 percentage points.
- Training Time: The training time of the model of `solov2_r50_fpn_1x` on Tesla v100 with 8 GPU is only 10 hours.