DJW c16313bb6a 第一次提交 | 9 месяцев назад | |
---|---|---|
.. | ||
README.md | 9 месяцев назад | |
mask-rcnn_r50-caffe_fpn_rpn-2conv_4conv1fc_syncbn-all_amp-lsj-100e_coco.py | 9 месяцев назад | |
mask-rcnn_r50-caffe_fpn_rpn-2conv_4conv1fc_syncbn-all_lsj-100e_coco.py | 9 месяцев назад | |
mask-rcnn_r50-caffe_fpn_rpn-2conv_4conv1fc_syncbn-all_lsj-400e_coco.py | 9 месяцев назад | |
mask-rcnn_r50_fpn_rpn-2conv_4conv1fc_syncbn-all_amp-lsj-100e_coco.py | 9 месяцев назад | |
mask-rcnn_r50_fpn_rpn-2conv_4conv1fc_syncbn-all_lsj-100e_coco.py | 9 месяцев назад | |
mask-rcnn_r50_fpn_rpn-2conv_4conv1fc_syncbn-all_lsj-50e_coco.py | 9 месяцев назад | |
metafile.yml | 9 месяцев назад |
We train Mask R-CNN with large-scale jitter and longer schedule as strong baselines. The modifications follow those in Detectron2.
Backbone | Style | Lr schd | Mem (GB) | Inf time (fps) | box AP | mask AP | Config | Download |
---|---|---|---|---|---|---|---|---|
R-50-FPN | pytorch | 50e | config | [model](<>) | [log](<>) | ||||
R-50-FPN | pytorch | 100e | config | [model](<>) | [log](<>) | ||||
R-50-FPN | caffe | 100e | 44.7 | 40.4 | config | [model](<>) | [log](<>) | ||
R-50-FPN | caffe | 400e | config | [model](<>) | [log](<>) |
When using large-scale jittering, there are sometimes empty proposals in the box and mask heads during training. This requires MMSyncBN that allows empty tensors. Therefore, please use mmcv-full>=1.3.14 to train models supported in this directory.