chegnyanjun

I have the same issue,do you know how to resolve?

最后在android上运行的时候提示这个问题，是这样加载库的，System.loadLibrary("squeezencnn");；libsqueezencnn.so放在main/libs/armeabi-v7a/下面，希望大佬的帮助啦。

@1s23 hi, I met the same problem, Do you have solved it now？

@rainbow520lxr,Thanks to get your reply,I have not solve the probelm because the tensorrt doesn't supports the nn.upsample() layer,Do you have any good ideas?Tje tensorrt6.0.1.5 seems not support write plugin layer...

@sunjiahao1999 ,hi,i met the same problem with Isaac-xie,and i check the mmdeploy/utils/constants.py file ,only find the MonocularDetection in Task，not in SDK_TASK_MAP,which it means only can run fcos3d in python ,...

i almost meet the same problem when i try to convert fcos3d pth model to trt model in the backtime,like this: (mmdeploy) mc@mc-System-Product-Name:/media/mc/autoDrivingData/works/deep_learning/autoDriving/MMDeploy$ python tools/deploy.py configs/mmdet3d/monocular-detection/monocular-detection_tensorrt_static-928x1600.py /home/mc/works/deep_learning/autoDriving/mmdetection3d/configs/fcos3d/fcos3d_r101_caffe_fpn_gn-head_dcn_2x8_1x_nus-mono3d.py /home/mc/works/deep_learning/autoDriving/mmdetection3d/checkpoints/fcos3d_r101_caffe_fpn_gn-head_dcn_2x8_1x_nus-mono3d_finetune_20210717_095645-8d806dc2.pth /home/mc/works/deep_learning/autoDriving/mmdetection3d/demo/data/nuscenes/n015-2018-07-24-11-22-45+0800__CAM_BACK__1532402927637525.jpg --work-dir...

好的，谢谢您啦。

这对应的是ssd网络的输出，你自己乘个倍数就可以得到矩形框，然后用opencv就可以画到原图了。