terminate called after throwing an instance of 'std::bad_alloc' what(): std::bad_alloc

[nullgogo]

Describe the bug

I'm trying to execute the preprocessing command on the nerfstudio documentation:

ns-process-data images --data data/nerfstudio/desolation/images --output-dir data/nerfstudio/desolation/colmap --no-gpu

But the following error appears:

` Copying images...[08:24:49] 🎉 Done copying images. process_data_utils.py:151 [08:26:31] 🎉 Done downscaling images. process_data_utils.py:213 [08:41:17] 🎉 Done extracting COLMAP features. colmap_utils.py:530 🚶 Running COLMAP feature matcher...──────────── 💀 💀 💀 ERROR 💀 💀 💀 ───────────── Error running command: colmap vocab_tree_matcher --database_path data/nerfstudio/desolation/colmap/colmap/database.db --SiftMatching.use_gpu 0 --VocabTreeMatching.vocab_tree_path /root/.local/share/nerfstudio/vocab_tree.fbow ─────────────────────────────────────────────────────────────────────────── terminate called after throwing an instance of 'std::bad_alloc'

what(): std::bad_alloc

*** Aborted at 1673340080 (unix time) try "date -d @1673340080" if you are using GNU date *** PC: @ 0x7f04fd85003b gsignal *** SIGABRT (@0x15bf) received by PID 5567 (TID 0x7f04f518d700) from PID 5567; stack trace: *** @ 0x7f0500124631 (unknown) @ 0x7f04ff7333c0 (unknown) @ 0x7f04fd85003b gsignal @ 0x7f04fd82f859 abort @ 0x7f04fdc4b911 (unknown) @ 0x7f04fdc5738c (unknown) @ 0x7f04fdc573f7 std::terminate() @ 0x7f04fdc576a9 __cxa_throw @ 0x7f04fdc4b522 (unknown) @ 0x55816a27e1c8 colmap::retrieval::VisualIndex<>::Read() @ 0x55816a356a9b colmap::VocabTreeFeatureMatcher::Run() @ 0x55816a3e6a9c colmap::Thread::RunFunc() @ 0x7f04fdc83de4 (unknown) @ 0x7f04ff727609 start_thread @ 0x7f04fd92c163 clone Aborted (core dumped)`

To Reproduce Installed colmap using the vcpkg succesfully. Then, run the command written above to preprocess custom data.

[tancik]

What do you try adding --matching-method exhaustive to your command to see if it makes a difference. If not, I'm not sure I can be of much help. This seems to be a colmap issue, you could try creating an issue in that repo. We do support some alternatives to colmap that you can also try - https://docs.nerf.studio/en/latest/quickstart/custom_dataset.html#using-custom-data

[tancik]

Closing due to inactivity. Feel free to reopen if it is still and issue.

[kirkding1]

Hi, I got the same problem and I also adding the code**--matching-method exhaustive**

It come out with this error:

Error running command: colmap exhaustive_matcher --database_path C:\Users\kirkd\OneDrive\Desktop\iCloud_Photos\statue_nav_nerf\colmap\database.db --SiftMatching.use_gpu 1

F0822 01:18:30.720218 43244 cache.h:132] Check failed: max_num_elems > 0 (0 vs. 0) *** Check failure stack trace: *** @ 00007FF9169D595A public: void __cdecl google::LogMessage::Flush(void) __ptr64 @ 00007FF9169D4539 public: __cdecl google::LogMessageFatal::~LogMessageFatal(void) __ptr64 @ 00007FF72E1CDE1D (unknown) @ 00007FF72E1DEC47 (unknown) @ 00007FF72E1D5D7D (unknown) @ 00007FF72E1C2130 (unknown) @ 00007FF72E1C095F (unknown) @ 00007FF92DF69363 _recalloc @ 00007FF92F4726AD BaseThreadInitThunk @ 00007FF9304EAA68 RtlUserThreadStart

@tancik

[maturk]

Hi, I got the same problem and I also adding the code**--matching-method exhaustive**

It come out with this error:

Error running command: colmap exhaustive_matcher --database_path C:\Users\kirkd\OneDrive\Desktop\iCloud_Photos\statue_nav_nerf\colmap\database.db --SiftMatching.use_gpu 1

F0822 01:18:30.720218 43244 cache.h:132] Check failed: max_num_elems > 0 (0 vs. 0) *** Check failure stack trace: *** @ 00007FF9169D595A public: void __cdecl google::LogMessage::Flush(void) __ptr64 @ 00007FF9169D4539 public: __cdecl google::LogMessageFatal::~LogMessageFatal(void) __ptr64 @ 00007FF72E1CDE1D (unknown) @ 00007FF72E1DEC47 (unknown) @ 00007FF72E1D5D7D (unknown) @ 00007FF72E1C2130 (unknown) @ 00007FF72E1C095F (unknown) @ 00007FF92DF69363 _recalloc @ 00007FF92F4726AD BaseThreadInitThunk @ 00007FF9304EAA68 RtlUserThreadStart

@tancik

Hi, there could be a problem with the colmap support for GPU with your colmap install. To test if this is the case, please try using the --no-gpu flag when running the ns-process-data command, and see if this does not fail.

[kirkding1]

Hi, I got the same problem and I also adding the code**--matching-method exhaustive** It come out with this error: Error running command: colmap exhaustive_matcher --database_path C:\Users\kirkd\OneDrive\Desktop\iCloud_Photos\statue_nav_nerf\colmap\database.db --SiftMatching.use_gpu 1 F0822 01:18:30.720218 43244 cache.h:132] Check failed: max_num_elems > 0 (0 vs. 0) *** Check failure stack trace: *** @ 00007FF9169D595A public: void __cdecl google::LogMessage::Flush(void) __ptr64 @ 00007FF9169D4539 public: __cdecl google::LogMessageFatal::~LogMessageFatal(void) __ptr64 @ 00007FF72E1CDE1D (unknown) @ 00007FF72E1DEC47 (unknown) @ 00007FF72E1D5D7D (unknown) @ 00007FF72E1C2130 (unknown) @ 00007FF72E1C095F (unknown) @ 00007FF92DF69363 _recalloc @ 00007FF92F4726AD BaseThreadInitThunk @ 00007FF9304EAA68 RtlUserThreadStart @tancik

Hi, there could be a problem with the colmap support for GPU with your colmap install. To test if this is the case, please try using the --no-gpu flag when running the ns-process-data command, and see if this does not fail.

It works now! Thank you!