ChaimZhu comments

Results 64 comments of


                                            ChaimZhu

DGCNN Trained Models on S3DIS Links are died

It's a little bit strange the old link is dead, we will update it soon.

DeepFusion implementation plans?

@jshilong Please have a look at this issue.

def inference_segmentor (model, pcd) -- intput & output

I'm not quite understand your questions, you mean, the structure of bin file or something else?

Support BEVDet series

We'll consider to support BEVDet in the future, besides, multi-camera 3D object detection is also the task we focus on.

Support BEVDet series

@Tai-Wang Please have a look at this to give some suggestions.

Will add FCOS-LiDAR? Thanks

According to the paper, FCOS-LiDAR is implemented based on MMDetection3D, once the code is open-source, we'll consider to migrate and add into our repo.

Use mmcv.runner.ModuleList instead of nn.ModuleList in mmdet3d.models.dense_heads.centerpoint_head

It's indeed a small bug in CenterPoint head, we'll fix it and re-train the model to see the performance improvement.

FCOS3D model conversion to onnx

I'm not quietly familiar with onnx, and the release of onnx conversion is currently not in our short-term plan. Maybe you can refer to this issue: https://github.com/open-mmlab/mmcv/issues/1422

FCOS3D model conversion to onnx

We do have a wechat group for mmdetection3d, but mainly talk about LiDAR based 3d object detection, I'm not sure whether someone could you help.

Can't visualize some data with MVX-Net model

We will fix the bug ASAP.