Messages

1

General / Re: Not enough video memory

« on: July 09, 2024, 07:39:00 AM »

OS: Windows 11 Pro
NVIDIA : GeForce GTX 660
NVIDIA Driver : 30.0.14.7444

After rebooting my coumter, It worked!
Thank you for your quick reply:)

2

General / Not enough video memory

« on: July 08, 2024, 01:45:06 PM »

Hi, Agisoft developers,

I met this issue: Not enough video memory.
The logs are below.
Please fix this issue!!

2024-07-08 19:41:07 Agisoft Metashape Professional Version: 2.1.2 build 18358 (64 bit)
2024-07-08 19:41:07 Platform: Windows
2024-07-08 19:41:07 CPU: Intel(R) Core(TM) i7-5930K CPU @ 3.50GHz (desktop)
2024-07-08 19:41:07 CPU family: 6 model: 63 signature: 306F2h
2024-07-08 19:41:07 RAM: 31.9 GB
2024-07-08 19:41:08 OpenGL Vendor: NVIDIA Corporation
2024-07-08 19:41:08 OpenGL Renderer: NVIDIA GeForce GTX 660/PCIe/SSE2
2024-07-08 19:41:08 OpenGL Version: 4.6.0 NVIDIA 474.44
2024-07-08 19:41:08 Maximum Texture Size: 16384
2024-07-08 19:41:08 Quad Buffered Stereo: not enabled
2024-07-08 19:41:08 ARB_vertex_buffer_object: supported
2024-07-08 19:41:08 ARB_texture_non_power_of_two: supported
2024-07-08 19:41:11 LoadProject: path = E:/Research_materials/Candidates/Vintage_DSM/2024Miho/2024MIHO_Vintage_DSM_1.psx
2024-07-08 19:41:11 loaded project in 0.003 sec
2024-07-08 19:41:12 Finished processing in 0.003 sec (exit code 1)
2024-07-08 19:41:18 AlignPhotos: downscale = 2, generic_preselection = on, reference_preselection = off, filter_mask = off, mask_tiepoints = off, filter_stationary_points = on, keypoint_limit = 40000, keypoint_limit_per_mpx = 1000, tiepoint_limit = 4000, guided_matching = off, exclude_corners = off, adaptive_fitting = on
2024-07-08 19:41:18 Matching photos...
2024-07-08 19:41:18 saved matching data in 0.006 sec
2024-07-08 19:41:18 scheduled 1 keypoint detection groups
2024-07-08 19:41:18 saved keypoint partition in 0.007 sec
2024-07-08 19:41:18 groups: 3
2024-07-08 19:41:18 3 of 3 used (100%)
2024-07-08 19:41:18 scheduled 1 keypoint matching groups
2024-07-08 19:41:18 saved matching partition in 0.004 sec
2024-07-08 19:41:18 loaded keypoint partition in 0 sec
2024-07-08 19:41:18 loaded matching data in 0 sec
2024-07-08 19:41:18 Detecting points...
2024-07-08 19:41:18 Found 1 GPUs in 0.044 sec (CUDA: 0.021 sec, OpenCL: 0.023 sec)
2024-07-08 19:41:18 Using device: NVIDIA GeForce GTX 660, 5 compute units, free memory: 1498/2048 MB, compute capability 3.0
2024-07-08 19:41:18   driver/runtime CUDA: 11040/10010
2024-07-08 19:41:18   max work group size 1024
2024-07-08 19:41:18   max work item sizes [1024, 1024, 64]
2024-07-08 19:41:19 [GPU] processing 16143x16143 image using 16143x8072 tiles
2024-07-08 19:41:20 Warning: cudaMemcpy(cuptr(), data, size, cudaMemcpyHostToDevice): out of memory (2) at line 244(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Error: cudaFree(data_): out of memory (2) at line 113(free memory unknown: out of memory (2))
2024-07-08 19:41:20 Warning: cudaStreamDestroy failed: out of memory (2)
2024-07-08 19:41:20 Finished processing in 2.581 sec (exit code 0)
2024-07-08 19:41:20 Error: cudaMemGetInfo(&free_mem_size, &total_mem_size): out of memory (2) at line 286(free memory unknown: out of memory (2))

3

General / Re: How to export variance of tie point?

« on: July 20, 2023, 05:18:19 AM »

Hi GGR,
Thank you for your reply. I will work on your suggestion.
Many thanks to you.

4

General / How to export variance of tie point?

« on: July 15, 2023, 10:13:53 AM »

Hello,

I would like to use the variance of the tie points to show the accuracy of DSM created by Meatashape.

How to export variance of tie point?

I hope to overlay the variance map of the tie points with exported DSM in the GIS program.

It is not easy to find the way of exporting it  in Metashape GUI. If you can give me any simple python program, I really appreciate it.

Furthermore, If the variance of the tie points is not the good idea in order to suggest the accuracy of the DSM, Which things are better?

Best,

5

General / Error in Build point cloud

« on: July 13, 2023, 04:29:08 AM »

Hello, 

Yesterday, there was no problem to handle the Metashape.
Today, It shows a error below with warning: Not enough memory.
I already restarted my computer, but I still get the same message.

2023-07-13 10:19:36 BuildPointCloud: quality = High, depth filtering = Mild, PM version, source data = Depth maps, point colors = 1
2023-07-13 10:19:36 Generating depth maps...
2023-07-13 10:19:36 Preparing 3 cameras info...
2023-07-13 10:19:36 cameras data loaded in 0.010388 s
2023-07-13 10:19:36 cameras graph built in 0.00019 s
2023-07-13 10:19:36 filtering neighbors with too low common points, threshold=50...
2023-07-13 10:19:36 avg neighbors before -> after filtering: 2 -> 0 (0% filtered out)
2023-07-13 10:19:36 limiting neighbors to 16 best...
2023-07-13 10:19:36 avg neighbors before -> after filtering: 2 -> 0 (0% filtered out)
2023-07-13 10:19:36 neighbors number min/1%/10%/median/90%/99%/max: 2, 2, 2, median=2, 2, 2, 2
2023-07-13 10:19:36 cameras info prepared in 0.011531 s
2023-07-13 10:19:36 saved cameras info in 0.009632
2023-07-13 10:19:36 Partitioning 3 cameras...
2023-07-13 10:19:36 number of mini clusters: 1
2023-07-13 10:19:36 1 groups: avg_ref=3 avg_neighb=0 total_io=100%
2023-07-13 10:19:36 max_ref=3 max_neighb=0 max_total=3
2023-07-13 10:19:36 cameras partitioned in 0.000114 s
2023-07-13 10:19:36 saved depth map partition in 0.002725 sec
2023-07-13 10:19:36 loaded cameras info in 0.000333
2023-07-13 10:19:36 loaded depth map partition in 1.7e-05 sec
2023-07-13 10:19:36 already partitioned (3<=50 ref cameras, 0<=200 neighb cameras)
2023-07-13 10:19:36 group 1/1: preparing 3 cameras images...
2023-07-13 10:19:36 tie points loaded in 0.00019 s
2023-07-13 10:19:36 Can't load OpenCL library
2023-07-13 10:19:36 Found 0 GPUs in 0.000643 sec (CUDA: 5e-06 sec, OpenCL: 0.000611 sec)
2023-07-13 10:19:36 Using device: CPU Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz (using 48/48)
2023-07-13 10:19:37 Finished processing in 0.646626 sec (exit code 0)
2023-07-13 10:19:37 Error: std::bad_alloc

How to fix this one?

Thanks

6

General / Question: Cameral model

« on: June 13, 2023, 07:30:38 AM »

Hi there,

I would like to create a DSM from three old airplane images, but I have no idea of their camera model.

Even though I do not know it, I can make a DSM from your program.

I usually click the button of the adaptive camera model fitting while doing align photos.

In doing so, the quality of DSM is poor.

How to enhance its quality?

If I insert the camera model and related information, Can get better result?
 
If I just approximate the camera model with another (like Wild RC5, RC8 (Aviogon Lens) or something in this site https://spymicmac.readthedocs.io/en/latest/tutorials/preprocessing/air_photos/cameras.html,

Can I improve my result?
 
Thank you for reading my post .

7

General / Re: Help me out: How to remove spike noise

« on: April 08, 2022, 10:22:14 AM »

Please help me out... Do you need more clear explanation? or information?

8

General / Help me out: How to remove spike noise

« on: April 07, 2022, 04:18:07 PM »

Dear Agisoft developers,

Before purchasing the Metashape, I tested this program in order to make a DEM from the historical aerial images with free-trial Metashape professional. These historical aerial photographs, captured in 1954 and 1968, do not have information about a focal length, a pixel size, and more. I met an obstacle, spike noise. I attached two screenshots. At_1 shows the historical DEM in the QGIS. At_2 indicates the historical DEM in the Metashape. As you can see the image in the Metashape, there are a number of spike noises disturbing the elevation analysis. How can I remove or suppress them?

I read the Metashape manual and found if I do not give the focal length and pixel size, the program assumes them as the certain values, leading to unexpected results.

Now I am waiting to process the DEM with highest accuracy without additional information.

If I do not have the focal length, the position of fiducial markers, pixel size, and so on, Couldn't I expect to remove or suppress the terrible spike noises? And Why does the noises occur?

Best regards

Changwan