Rtakaha

Results 11 comments of Rtakaha

Criterion to determine the x-y plane offset

I think NeuralRecon cannot generate mesh in the area z < 0. From the [comment](https://github.com/zju3dv/NeuralRecon/blob/cd047e2356f68f60adfb923f33572d712f64b57a/tools/process_arkit_data.py#L83), I assume this is because ScanNet has mesh data only in the area z >=...

adding texture to the meshes generated for the demo data

This [comment](https://github.com/zju3dv/NeuralRecon/issues/27#issuecomment-874031869) from #27 might be helpful for you. https://github.com/nmoehrle/mvs-texturing

python3 train.py --cfg_file cfgs/kitti_models/VFF_PVRCNN.yaml

I fixed this problem by reshaping. ```python points_depth = points_t[..., -1] - raw_project[..., 2, 3].reshape(points_t.shape[0], 1) ```

Add `to(device)` to avoid device missmatch error.

I needed to add `to(device)` here when I use GPU. `images[i_test]` is defined on GPU, but `torch.Tensor(rgbs)` is defined on CPU without `to(device)`. ```python # run_nerf.py # test_loss = img2mse(torch.Tensor(rgbs),...

Cannot run trainer.py with --model=car.waymo_deepfusion.DeepFusionCenterPointPed, undefined symbol: _ZNK10tensorflow8OpKernel11TraceStringERKNS_15OpKernelContextEb

Same error occurs when I try `--model=car.waymo.StarNetVehicle`, yet `--model=car.kitti.StarNetCarModel0701` does not produce this error.

Cannot run trainer.py with --model=car.waymo_deepfusion.DeepFusionCenterPointPed, undefined symbol: _ZNK10tensorflow8OpKernel11TraceStringERKNS_15OpKernelContextEb

The output of `bazel build` after downgrading tensorflow to `tensorflow==2.6.0`. ```txt Extracting Bazel installation... Starting local Bazel server and connecting to it... DEBUG: Rule 'subpar' indicated that a canonical reproducible...

Cannot run trainer.py with --model=car.waymo_deepfusion.DeepFusionCenterPointPed, undefined symbol: _ZNK10tensorflow8OpKernel11TraceStringERKNS_15OpKernelContextEb

tensorflow is compiled with `_GLIBCXX_USE_CXX11_ABI=1` from 2.9.0, and I believe this causes the problem. https://github.com/tensorflow/tensorflow/releases/tag/v2.9.0 Tried below with tensorflow=2.7.3, but didn't work (build failed, same error as the [previous comment...

Cannot run trainer.py with --model=car.waymo_deepfusion.DeepFusionCenterPointPed, undefined symbol: _ZNK10tensorflow8OpKernel11TraceStringERKNS_15OpKernelContextEb

I have finally fixed this by setting back `_GLIBCXX_USE_CXX11_ABI` to `0` in several files and compile with tensorflow==2.7.3! Let me summary this issue for my memorandum and for people who...

about data augmentation

Thank you very much for sharing your work, @happinesslz ! I have the same question. Data augmentation is applied [here](https://github.com/happinesslz/EPNet/blob/master/lib/datasets/kitti_rcnn_dataset.py#L381-L385), and `AUG_DATA` is `True` when the given config is used....

about data augmentation

Sorry that I didn't check previous issues enough. Now I understand how img and pcd are aligned. As [this comment](https://github.com/happinesslz/EPNet/issues/22#issuecomment-821823302) mentions, pcd and img are aligned [here](https://github.com/happinesslz/EPNet/blob/master/lib/datasets/kitti_rcnn_dataset.py#L310), by storing the...