1
General / Re: "Unexpected Channel Count" error
« on: January 27, 2019, 12:53:59 AM »
Email is sent.
Mike
Mike
Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Pages: [1] 2
2
General / Re: "Unexpected Channel Count" error
« on: January 26, 2019, 07:48:34 PM »
I also experienced the "Unexpected channel count" error using 1.5.0 build 7492.
This happened on a project of 216 TIFF images. 205 were masked in Photoshop so they included an alpha channel. The remaining 11 images were not masked and included no alpha channel. Photos aligned as usual and a good sparse cloud was created. The "Unexpected channel count" error occurred at the end of Dense Cloud creation.
Although no dense cloud was created, I noticed a depth map had been. I don't use depth maps but I also noticed the Mesh panel said a mesh could be created from the depth map. I tried that, and it worked even though the error occurred for the dense cloud. The Shaded, Solid, and Wireframe models looked fine.
Next, I proceeded to create a textured model. Again, the "Unexpected channel count" error occurred. This time, the Console indicated 11 images contained only 3 channels rather than 4. Those images were then disabled, and a good textured model was created. Still, no dense cloud was ever created.
I plan to create masks for those 11 images that originally didn't have them, and rerun the project with all images having masks.
Before the update to 1.5.0, I never had a problem mixing masked and unmasked TIFF images.
From the first texture attempt:
2019-01-26 06:24:55 - 100% done by [CPU 1] with timings: 2267 s = 0% fetch + 1% init + 77% hist (96% GPU) + 0% init + 21% iters (81% + 19%) + 0% updating
2019-01-26 06:24:55 IO utilization: 1%
2019-01-26 06:24:55 Level done in 2268.3 sec
2019-01-26 06:24:55 Surface estimated in 2454.16 sec
2019-01-26 06:24:55 Extracting surface...
2019-01-26 06:26:06 8513494 vertices total
2019-01-26 06:26:06 17026857 faces total
2019-01-26 06:26:06 done in 71.457 sec
2019-01-26 06:26:06 Calculating colors...
2019-01-26 06:26:09 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:09 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:09 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:10 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:10 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:12 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:12 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:13 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:13 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:13 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:15 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:28:20 done in 133.252 sec
2019-01-26 06:28:20 Decimating surface...
2019-01-26 06:28:35 decimated to 479561 faces
2019-01-26 06:28:35 done in 15.277 sec
2019-01-26 06:28:36 Finished processing in 29467.4 sec (exit code 1)
This happened on a project of 216 TIFF images. 205 were masked in Photoshop so they included an alpha channel. The remaining 11 images were not masked and included no alpha channel. Photos aligned as usual and a good sparse cloud was created. The "Unexpected channel count" error occurred at the end of Dense Cloud creation.
Although no dense cloud was created, I noticed a depth map had been. I don't use depth maps but I also noticed the Mesh panel said a mesh could be created from the depth map. I tried that, and it worked even though the error occurred for the dense cloud. The Shaded, Solid, and Wireframe models looked fine.
Next, I proceeded to create a textured model. Again, the "Unexpected channel count" error occurred. This time, the Console indicated 11 images contained only 3 channels rather than 4. Those images were then disabled, and a good textured model was created. Still, no dense cloud was ever created.
I plan to create masks for those 11 images that originally didn't have them, and rerun the project with all images having masks.
Before the update to 1.5.0, I never had a problem mixing masked and unmasked TIFF images.
From the first texture attempt:
2019-01-26 06:24:55 - 100% done by [CPU 1] with timings: 2267 s = 0% fetch + 1% init + 77% hist (96% GPU) + 0% init + 21% iters (81% + 19%) + 0% updating
2019-01-26 06:24:55 IO utilization: 1%
2019-01-26 06:24:55 Level done in 2268.3 sec
2019-01-26 06:24:55 Surface estimated in 2454.16 sec
2019-01-26 06:24:55 Extracting surface...
2019-01-26 06:26:06 8513494 vertices total
2019-01-26 06:26:06 17026857 faces total
2019-01-26 06:26:06 done in 71.457 sec
2019-01-26 06:26:06 Calculating colors...
2019-01-26 06:26:09 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:09 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:09 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:10 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:10 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:12 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:12 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:13 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:13 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:13 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:26:15 Warning! Image ignored because it has 3 channels, but 4 channels expected!
2019-01-26 06:28:20 done in 133.252 sec
2019-01-26 06:28:20 Decimating surface...
2019-01-26 06:28:35 decimated to 479561 faces
2019-01-26 06:28:35 done in 15.277 sec
2019-01-26 06:28:36 Finished processing in 29467.4 sec (exit code 1)
3
General / Re: focus stacking for photogrammetry
« on: December 03, 2018, 07:54:55 AM »
If you are still developing your workflow for focus stacking, then you might glance over this article:
COUPLING FOCUS STACKING WITH PHOTOGRAMMETRY TO ILLUSTRATE SMALL FOSSIL TEETH, https://www.jpaleontologicaltechniques.org/pasta3/JPT%20N18/Bulletin.html
COUPLING FOCUS STACKING WITH PHOTOGRAMMETRY TO ILLUSTRATE SMALL FOSSIL TEETH, https://www.jpaleontologicaltechniques.org/pasta3/JPT%20N18/Bulletin.html
4
General / Re: Tilt Shift Lens vs Focus Stacking
« on: November 26, 2018, 05:49:05 AM »
Focus stacking can be a lot of work. It takes some patience but it can be somewhat automated. Anyway, it will work with photogrammetry projects.
https://www.jpaleontologicaltechniques.org/pasta3/JPT%20N18/Bulletin.html
https://www.jpaleontologicaltechniques.org/pasta3/JPT%20N18/Bulletin.html
5
Bug Reports / Re: photos do not display properly when double clicked
« on: July 15, 2018, 09:47:40 AM »
Alex,
Thank you. I was using the shortcut made from the Windows Start menu. It would not let me enter anything into the Target box. Then I created a shortcut directly by clicking on photoscan.exe. That shortcut allowed me to enter the command you previously posted. Now photos are viewable in the usual way.
Mike
Thank you. I was using the shortcut made from the Windows Start menu. It would not let me enter anything into the Target box. Then I created a shortcut directly by clicking on photoscan.exe. That shortcut allowed me to enter the command you previously posted. Now photos are viewable in the usual way.
Mike
6
Bug Reports / Re: photos do not display properly when double clicked
« on: July 15, 2018, 12:33:22 AM »
Alex,
I copied the opengl angle command from your previous message and pasted it into the properties panel for the shortcut. I also update to the version 1.4.3, but the behavior is the same.
Also attached is a screenshot showing the Console after trying to display a photo.
Mike
I copied the opengl angle command from your previous message and pasted it into the properties panel for the shortcut. I also update to the version 1.4.3, but the behavior is the same.
Also attached is a screenshot showing the Console after trying to display a photo.
Mike
7
Bug Reports / Re: photos do not display properly when double clicked
« on: July 13, 2018, 07:52:02 PM »
I am having the same problem with displaying photos. Double-clicking an image file in either the Workspace or Photos pane displays only a blank image, but with the image file name correctly included on the tab.
My shortcut was modified as Alex suggested, but that has not corrected the problem.
I'm using the latest build, v.1.4.2 (6205).
My shortcut was modified as Alex suggested, but that has not corrected the problem.
I'm using the latest build, v.1.4.2 (6205).
8
General / Re: Incompatible dense cloud colors
« on: April 25, 2018, 03:32:27 PM »
My usual workflow involves recording RAW images that are later converted to 8 bit TIFF files.
The updates included with Version 1.4.1 build 5925 (26 February 2018) resolved the "Incompatible dense cloud colors" issue for me. I haven't updated yet to the latest build.
Mike
The updates included with Version 1.4.1 build 5925 (26 February 2018) resolved the "Incompatible dense cloud colors" issue for me. I haven't updated yet to the latest build.
Mike
9
General / Re: Incompatible dense cloud colors
« on: January 22, 2018, 05:07:04 AM »
Alexey,
As you suggested, disabling the integrated graphics card let the processing complete. Thanks again.
Just to clarify the situation, is it the case that the issue about sparse/dense cloud color error will be resolved in the next update?
Mike
As you suggested, disabling the integrated graphics card let the processing complete. Thanks again.
Just to clarify the situation, is it the case that the issue about sparse/dense cloud color error will be resolved in the next update?
Mike
10
General / Re: Incompatible dense cloud colors
« on: January 19, 2018, 02:09:20 AM »
Alexey,
Just to check things out I began a completely new project file with just one of my original chunks of 12 images. Here is the complete display from the console window:
2018-01-18 08:35:09 Agisoft PhotoScan Professional Version: 1.4.0 build 5650 (64 bit)
2018-01-18 08:35:09 Platform: Windows
2018-01-18 08:35:09 CPU: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz (laptop)
2018-01-18 08:35:09 CPU family: 6 model: 94 signature: 506E3h
2018-01-18 08:35:09 RAM: 31.8 GB
2018-01-18 08:35:10 OpenGL Vendor: NVIDIA Corporation
2018-01-18 08:35:10 OpenGL Renderer: GeForce GTX 965M/PCIe/SSE2
2018-01-18 08:35:10 OpenGL Version: 4.6.0 NVIDIA 390.65
2018-01-18 08:35:10 Maximum Texture Size: 16384
2018-01-18 08:35:10 Quad Buffered Stereo: not enabled
2018-01-18 08:35:10 ARB_vertex_buffer_object: supported
2018-01-18 08:35:10 ARB_texture_non_power_of_two: supported
2018-01-18 08:35:10 Using simple console. Rich console can be enabled in Preferences dialog.
2018-01-18 08:35:50 Loading photos...
2018-01-18 08:35:50 Finished processing in 0.015 sec (exit code 1)
2018-01-18 08:36:58 AlignPhotos: accuracy = High, preselection = none, keypoint limit = 60000, tiepoint limit = 0, constrain features by mask = 0, adaptive fitting = 0
2018-01-18 08:36:58 Matching photos...
2018-01-18 08:36:58 Detecting points...
2018-01-18 08:36:59 Using device: GeForce GTX 965M, 8 compute units, 4096 MB global memory, compute capability 5.2
2018-01-18 08:36:59 driver version: 9010, runtime version: 5050
2018-01-18 08:36:59 max work group size 1024
2018-01-18 08:36:59 max work item sizes [1024, 1024, 64]
2018-01-18 08:36:59 Using device: Intel(R) HD Graphics 530, 24 compute units, 13024 MB global memory, OpenCL 2.0
2018-01-18 08:36:59 driver version: 22.20.16.4749, platform version: OpenCL 2.0
2018-01-18 08:36:59 max work group size 256
2018-01-18 08:36:59 max work item sizes [256, 256, 256]
2018-01-18 08:36:59 max mem alloc size 2047 MB
2018-01-18 08:36:59 Building kernels for Intel(R) HD Graphics 530...
2018-01-18 08:37:01 [CPU] photo 0: 60000 points
2018-01-18 08:37:01 [GPU] photo 1: 60000 points
2018-01-18 08:37:03 Kernels compilation done in 3.725 seconds
2018-01-18 08:37:03 [GPU] photo 4: 60000 points
2018-01-18 08:37:04 [CPU] photo 3: 60000 points
2018-01-18 08:37:04 Building kernels for Intel(R) HD Graphics 530...
2018-01-18 08:37:04 Kernels compilation done in 0.407 seconds
2018-01-18 08:37:05 [GPU] photo 2: 60000 points
2018-01-18 08:37:05 [GPU] photo 5: 60000 points
2018-01-18 08:37:06 [CPU] photo 6: 60000 points
2018-01-18 08:37:06 [GPU] photo 7: 60000 points
2018-01-18 08:37:07 [GPU] photo 8: 60000 points
2018-01-18 08:37:08 [GPU] photo 10: 60000 points
2018-01-18 08:37:09 [CPU] photo 9: 60000 points
2018-01-18 08:37:09 [GPU] photo 11: 60000 points
2018-01-18 08:37:09 points detected in 11.293 sec
2018-01-18 08:37:10 Matching points...
2018-01-18 08:37:10 Using device: GeForce GTX 965M, 8 compute units, 4096 MB global memory, compute capability 5.2
2018-01-18 08:37:10 driver version: 9010, runtime version: 5050
2018-01-18 08:37:10 max work group size 1024
2018-01-18 08:37:10 max work item sizes [1024, 1024, 64]
2018-01-18 08:37:10 Using device: Intel(R) HD Graphics 530, 24 compute units, 13024 MB global memory, OpenCL 2.0
2018-01-18 08:37:10 driver version: 22.20.16.4749, platform version: OpenCL 2.0
2018-01-18 08:37:10 max work group size 256
2018-01-18 08:37:10 max work item sizes [256, 256, 256]
2018-01-18 08:37:10 max mem alloc size 2047 MB
2018-01-18 08:57:57 Finished processing in 1258.84 sec (exit code 0)
2018-01-18 08:57:57 Error: Kernel matchBruteForce_Hamming: CL_UNKNOWN_ERROR_CODE_-14 (-14) at line 374
>>>
Just to check things out I began a completely new project file with just one of my original chunks of 12 images. Here is the complete display from the console window:
2018-01-18 08:35:09 Agisoft PhotoScan Professional Version: 1.4.0 build 5650 (64 bit)
2018-01-18 08:35:09 Platform: Windows
2018-01-18 08:35:09 CPU: Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz (laptop)
2018-01-18 08:35:09 CPU family: 6 model: 94 signature: 506E3h
2018-01-18 08:35:09 RAM: 31.8 GB
2018-01-18 08:35:10 OpenGL Vendor: NVIDIA Corporation
2018-01-18 08:35:10 OpenGL Renderer: GeForce GTX 965M/PCIe/SSE2
2018-01-18 08:35:10 OpenGL Version: 4.6.0 NVIDIA 390.65
2018-01-18 08:35:10 Maximum Texture Size: 16384
2018-01-18 08:35:10 Quad Buffered Stereo: not enabled
2018-01-18 08:35:10 ARB_vertex_buffer_object: supported
2018-01-18 08:35:10 ARB_texture_non_power_of_two: supported
2018-01-18 08:35:10 Using simple console. Rich console can be enabled in Preferences dialog.
2018-01-18 08:35:50 Loading photos...
2018-01-18 08:35:50 Finished processing in 0.015 sec (exit code 1)
2018-01-18 08:36:58 AlignPhotos: accuracy = High, preselection = none, keypoint limit = 60000, tiepoint limit = 0, constrain features by mask = 0, adaptive fitting = 0
2018-01-18 08:36:58 Matching photos...
2018-01-18 08:36:58 Detecting points...
2018-01-18 08:36:59 Using device: GeForce GTX 965M, 8 compute units, 4096 MB global memory, compute capability 5.2
2018-01-18 08:36:59 driver version: 9010, runtime version: 5050
2018-01-18 08:36:59 max work group size 1024
2018-01-18 08:36:59 max work item sizes [1024, 1024, 64]
2018-01-18 08:36:59 Using device: Intel(R) HD Graphics 530, 24 compute units, 13024 MB global memory, OpenCL 2.0
2018-01-18 08:36:59 driver version: 22.20.16.4749, platform version: OpenCL 2.0
2018-01-18 08:36:59 max work group size 256
2018-01-18 08:36:59 max work item sizes [256, 256, 256]
2018-01-18 08:36:59 max mem alloc size 2047 MB
2018-01-18 08:36:59 Building kernels for Intel(R) HD Graphics 530...
2018-01-18 08:37:01 [CPU] photo 0: 60000 points
2018-01-18 08:37:01 [GPU] photo 1: 60000 points
2018-01-18 08:37:03 Kernels compilation done in 3.725 seconds
2018-01-18 08:37:03 [GPU] photo 4: 60000 points
2018-01-18 08:37:04 [CPU] photo 3: 60000 points
2018-01-18 08:37:04 Building kernels for Intel(R) HD Graphics 530...
2018-01-18 08:37:04 Kernels compilation done in 0.407 seconds
2018-01-18 08:37:05 [GPU] photo 2: 60000 points
2018-01-18 08:37:05 [GPU] photo 5: 60000 points
2018-01-18 08:37:06 [CPU] photo 6: 60000 points
2018-01-18 08:37:06 [GPU] photo 7: 60000 points
2018-01-18 08:37:07 [GPU] photo 8: 60000 points
2018-01-18 08:37:08 [GPU] photo 10: 60000 points
2018-01-18 08:37:09 [CPU] photo 9: 60000 points
2018-01-18 08:37:09 [GPU] photo 11: 60000 points
2018-01-18 08:37:09 points detected in 11.293 sec
2018-01-18 08:37:10 Matching points...
2018-01-18 08:37:10 Using device: GeForce GTX 965M, 8 compute units, 4096 MB global memory, compute capability 5.2
2018-01-18 08:37:10 driver version: 9010, runtime version: 5050
2018-01-18 08:37:10 max work group size 1024
2018-01-18 08:37:10 max work item sizes [1024, 1024, 64]
2018-01-18 08:37:10 Using device: Intel(R) HD Graphics 530, 24 compute units, 13024 MB global memory, OpenCL 2.0
2018-01-18 08:37:10 driver version: 22.20.16.4749, platform version: OpenCL 2.0
2018-01-18 08:37:10 max work group size 256
2018-01-18 08:37:10 max work item sizes [256, 256, 256]
2018-01-18 08:37:10 max mem alloc size 2047 MB
2018-01-18 08:57:57 Finished processing in 1258.84 sec (exit code 0)
2018-01-18 08:57:57 Error: Kernel matchBruteForce_Hamming: CL_UNKNOWN_ERROR_CODE_-14 (-14) at line 374
>>>
11
General / Re: Incompatible dense cloud colors
« on: January 18, 2018, 07:01:29 AM »
Alexey,
I downloaded the file you made for me and tried to continue processing by aligning the photos of the merged chunk with reset alignment checked. It ran for a while and then issued this message:
2018-01-17 19:21:24 Error: Kernel matchBruteForce_Hamming: CL_UNKNOWN_ERROR_CODE_-14 (-14) at line 374
Also, with version 1.4.0 installed I no longer see Gradual Selection on the Edit menu, and I don't see anything about it in the Change Log. Has the Gradual Selection been removed or replaced by something else?
Mike
I downloaded the file you made for me and tried to continue processing by aligning the photos of the merged chunk with reset alignment checked. It ran for a while and then issued this message:
2018-01-17 19:21:24 Error: Kernel matchBruteForce_Hamming: CL_UNKNOWN_ERROR_CODE_-14 (-14) at line 374
Also, with version 1.4.0 installed I no longer see Gradual Selection on the Edit menu, and I don't see anything about it in the Change Log. Has the Gradual Selection been removed or replaced by something else?
Mike
12
General / Re: Incompatible dense cloud colors
« on: January 16, 2018, 07:12:26 PM »
Alexey,
There is no time pressure for completing this project, but if you don't mind sending me your file that would be great. Otherwise, I don't mind waiting for the update.
Your help is greatly appreciated.
Mike
There is no time pressure for completing this project, but if you don't mind sending me your file that would be great. Otherwise, I don't mind waiting for the update.
Your help is greatly appreciated.
Mike
13
General / Re: Incompatible dense cloud colors
« on: January 16, 2018, 04:14:02 AM »
Alexey,
I installed version 1.4.0, opened my file, and tried to merge the chunks. The same error message occurred. You can get my project and image files here:
https://www.dropbox.com/sh/ve3cbj7y59y4q51/AAD3v-jJGdsPE2-D7cFqnuXJa?dl=0
The workflow I'm using worked on about 10 other close-range projects without any problems, but the last one I did was about 10 months ago. I'm using the same computer and camera as before. Now, I'm planning to do many more close-range projects so I hope you can help me get this resolved. Thanks
I installed version 1.4.0, opened my file, and tried to merge the chunks. The same error message occurred. You can get my project and image files here:
https://www.dropbox.com/sh/ve3cbj7y59y4q51/AAD3v-jJGdsPE2-D7cFqnuXJa?dl=0
The workflow I'm using worked on about 10 other close-range projects without any problems, but the last one I did was about 10 months ago. I'm using the same computer and camera as before. Now, I'm planning to do many more close-range projects so I hope you can help me get this resolved. Thanks
14
General / Re: Incompatible dense cloud colors
« on: January 15, 2018, 08:37:52 AM »
Alexey,
My usual workflow with more than one chunk is to align and merge the chunks before further processing. Today, I processed my three chunks into the dense clouds. The three dense cloud chunks were aligned and look great. Once again, when I tried merging the three dense cloud chunks, this was reported in the console window:
2018-01-14 21:12:02 Dense clouds have different color format, they are merged as RGB
2018-01-14 21:12:03 Finished processing in 10.834 sec (exit code 0)
2018-01-14 21:12:03 Error: Incompatible dense cloud colors
All my photos were shot in RAW format and converted to TIFFs. The same procedure was used to process all the images. No attempt was made to alter the format of the images. They are all color images. Does the dense cloud color format have any relationship to the image format? Would it matter if the images were converted to a format different than TIFF?
My usual workflow with more than one chunk is to align and merge the chunks before further processing. Today, I processed my three chunks into the dense clouds. The three dense cloud chunks were aligned and look great. Once again, when I tried merging the three dense cloud chunks, this was reported in the console window:
2018-01-14 21:12:02 Dense clouds have different color format, they are merged as RGB
2018-01-14 21:12:03 Finished processing in 10.834 sec (exit code 0)
2018-01-14 21:12:03 Error: Incompatible dense cloud colors
All my photos were shot in RAW format and converted to TIFFs. The same procedure was used to process all the images. No attempt was made to alter the format of the images. They are all color images. Does the dense cloud color format have any relationship to the image format? Would it matter if the images were converted to a format different than TIFF?
15
General / Re: Incompatible dense cloud colors
« on: January 14, 2018, 08:49:25 AM »
Alexey,
I am using v 1.3.4 build 5067 (64 bit). The chunks were processed in this version on the same computer.
I am using v 1.3.4 build 5067 (64 bit). The chunks were processed in this version on the same computer.
Pages: [1] 2
