cdcseacave / openMVS

open Multi-View Stereo reconstruction library
http://cdcseacave.github.io
GNU Affero General Public License v3.0
3.26k stars 898 forks source link

DensifyPointCloud inv_scale_x > 0 in function 'resize' issue #949

Open WangYu0611 opened 1 year ago

WangYu0611 commented 1 year ago

16:20:42 [App ] Build date: Mar 5 2023, 15:54:51 16:20:42 [App ] CPU: 12th Gen Intel(R) Core(TM) i7-12700KF (16 cores) 16:20:42 [App ] RAM: 38.55GB Physical Memory 2.00GB Virtual Memory 16:20:42 [App ] OS: Linux 5.19.0-35-generic (x86_64) 16:20:42 [App ] Disk: 135.81GB (195.31GB) space 16:20:42 [App ] SSE & AVX compatible CPU & OS detected 16:20:42 [App ] Command line: DensifyPointCloud scene.mvs -o depthData 16:20:42 [App ] Camera model loaded: platform 0; camera 0; f 0.422x0.422; poses 1162 16:20:42 [App ] Scene loaded from interface format (94ms): 1162 images (1162 calibrated) with a total of 9191.60 MPixels (7.91 MPixels/image) 426067 points, 0 vertices, 0 faces 16:20:42 [App ] Found a camera not pointing towards the scene center; the scene will be considered unbounded (no ROI) 16:20:42 [App ] Point-cloud composed of 426067 points with:

Aborted (core dumped)

WangYu0611 commented 1 year ago

https://drive.google.com/file/d/1ahd9jf4CTrBsXs68_Ou0U5k7eRdyJGjf/view?usp=share_link

data link : https://drive.google.com/file/d/1ahd9jf4CTrBsXs68_Ou0U5k7eRdyJGjf/view?usp=share_link

WangYu0611 commented 1 year ago

@cdcseacave can you help me

cdcseacave commented 1 year ago

Pls build openMVS in release mode

On Thu, 9 Mar 2023 at 04:20 WangYu0611 @.***> wrote:

@cdcseacave https://github.com/cdcseacave can you help me

— Reply to this email directly, view it on GitHub https://github.com/cdcseacave/openMVS/issues/949#issuecomment-1461168793, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU . You are receiving this because you were mentioned.Message ID: @.***>

WangYu0611 commented 1 year ago

Pls build openMVS in release mode On Thu, 9 Mar 2023 at 04:20 WangYu0611 @.> wrote: @cdcseacave https://github.com/cdcseacave can you help me — Reply to this email directly, view it on GitHub <#949 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU . You are receiving this because you were mentioned.Message ID: @.>

I'll try it right away

WangYu0611 commented 1 year ago

请在发布模式下构建 openMVS …… 2023 年 3 月 9 日星期四 04:20 WangYu0611 @.> 写道: @cdcseacave < https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe-auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到此邮件是因为有人提到了您。邮件 ID:@>

I downloaded the Nov 28, 2022 version of the code from the Releases on the right side of this address: https://github.com/cdcseacave/openMVS, and then recompiled with the make command, the same error will still be reported. Sorry, I am A new student who is just using this library, there may be something that is not standard. So this is what I mean by rebuilding releases?

error : 16:09:17 [App ] Build date: Mar 9 2023, 16:07:01 16:09:17 [App ] CPU: 12th Gen Intel(R) Core(TM) i7-12700KF (16 cores) 16:09:17 [App ] RAM: 38.55GB Physical Memory 2.00GB Virtual Memory 16:09:17 [App ] OS: Linux 5.19.0-35-generic (x86_64) 16:09:17 [App ] Disk: 117.81GB (195.31GB) space 16:09:17 [App ] SSE & AVX compatible CPU & OS detected 16:09:17 [App ] Command line: DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App ] Camera model loaded: platform 0; camera 0; f 0.422x0.422; poses 1162 16:09:17 [App ] Scene loaded from interface format (109ms): 1162 images (1162 calibrated) with a total of 9191.60 MPixels (7.91 MPixels/image) 426067 points, 0 vertices, 0 faces 16:09:17 [App ] Found a camera not pointing towards the scene center; the scene will be considered unbounded (no ROI) 16:09:17 [App ] Point-cloud composed of 426067 points with:

Aborted (core dumped)

wutingfan commented 1 year ago

Pls build openMVS in release mode On Thu, 9 Mar 2023 at 04:20 WangYu0611 @.> wrote: @cdcseacave https://github.com/cdcseacave can you help me — Reply to this email directly, view it on GitHub <#949 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU . You are receiving this because you were mentioned.Message ID: @.>

I believe it's a bug, I meet the same problem when using cpu estimate depth-maps, not always happen, but It could be. I'm doing some tests to find the problem.

wutingfan commented 1 year ago

请在发布模式下构建 openMVS …… 2023 年 3 月 9 日星期四 04:20 WangYu0611 @.> 写道: @cdcseacave < https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe-auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到此邮件是因为有人提到了您。邮件 ID:@>

I downloaded the Nov 28, 2022 version of the code from the Releases on the right side of this address: https://github.com/cdcseacave/openMVS, and then recompiled with the make command, the same error will still be reported. Sorry, I am A new student who is just using this library, there may be something that is not standard. So this is what I mean by rebuilding releases?

error : 16:09:17 [App ] Build date: Mar 9 2023, 16:07:01 16:09:17 [App ] CPU: 12th Gen Intel(R) Core(TM) i7-12700KF (16 cores) 16:09:17 [App ] RAM: 38.55GB Physical Memory 2.00GB Virtual Memory 16:09:17 [App ] OS: Linux 5.19.0-35-generic (x86_64) 16:09:17 [App ] Disk: 117.81GB (195.31GB) space 16:09:17 [App ] SSE & AVX compatible CPU & OS detected 16:09:17 [App ] Command line: DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App ] Camera model loaded: platform 0; camera 0; f 0.422x0.422; poses 1162 16:09:17 [App ] Scene loaded from interface format (109ms): 1162 images (1162 calibrated) with a total of 9191.60 MPixels (7.91 MPixels/image) 426067 points, 0 vertices, 0 faces 16:09:17 [App ] Found a camera not pointing towards the scene center; the scene will be considered unbounded (no ROI) 16:09:17 [App ] Point-cloud composed of 426067 points with:

  • visibility info (1477808 views - 3.47 views/point): 0 points with 1- views (0.00%) 254544 points with 2 views (59.74%) 73345 points with 3 views (17.21%) 98178 points with 4+ views (23.04%) 2 min / 3.46849 mean (3.98363 std) / 109 max 16:09:24 [App ] Preparing images for dense reconstruction completed: 1162 images (6s687ms) 16:09:25 [App ] Selecting images for dense reconstruction completed: 1162 images (797ms) Estimated depth-maps 111 (9.55%, 11m6s, ETA 1h45m)...terminate called after throwing an instance of 'cv::Exception' what(): OpenCV(4.5.4) ./modules/imgproc/src/resize.cpp:4054: error: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

Aborted (core dumped)

I meet the same problem, did you solev it?

WANGYU312q commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese?

I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

wutingfan commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese?

I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

WANGYU312q commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese? I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

我用小米12Pro拍摄的照片都可以正常计算

WANGYU312q commented 1 year ago

应该是用的CPU计算的

wutingfan commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese? I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

我用小米12Pro拍摄的照片都可以正常计算

之前出错的数据还有吗,上面的链接失效了

WangYu0611 commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese? I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

我用小米12Pro拍摄的照片都可以正常计算

之前出错的数据还有吗,上面的链接失效了

没有了,我的谷歌网盘不够就删掉了,上面那段数据是会出错, 对你应该没有任何帮助。

wutingfan commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese? I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

我用小米12Pro拍摄的照片都可以正常计算

之前出错的数据还有吗,上面的链接失效了

没有了,我的谷歌网盘不够就删掉了,上面那段数据是会出错, 对你应该没有任何帮助。

我就是想要复现这个问题,不然不好查bug,再检查一下你的运行环境,看一下还有没有没删除的,你的问题怎么解决的?后来没再出现过这个问题?

wutingfan commented 1 year ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese? I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

我用小米12Pro拍摄的照片都可以正常计算

之前出错的数据还有吗,上面的链接失效了

没有了,我的谷歌网盘不够就删掉了,上面那段数据是会出错, 对你应该没有任何帮助。

给一下你的邮箱,可以吗?这样效率比较低

hai-bo1017 commented 1 month ago

了,我的谷歌网盘不够就删掉了,上面那段数据是会出错, 对你应该没有任何帮助。

给一下你的邮箱,可以吗?这样效率比较低

have you solved this problem?

silkmeloni commented 1 month ago

请在发布模式下构建openMVS …… 2023年3月9日星期四04:20 WangYu0611 @ . > 写道:@cdcseacave< https://github.com/cdcseacave > can you help me — 直接回复此邮件,查看它在 GitHub < #949 (comment) > 上,或取消订阅 < https://github.com/notifications/unsubscribe- auth/AAVMH3W5QL2SBNYGST3SR4TW3E47JANCNFSM6AAAAAAVTO5YSU >。您收到这个邮件是因为有人提了您。邮件ID: @ >

我从这个地址右侧的Releases下载了Nov 28, 2022版本的代码:https://github.com/cdcseacave/openMVS,然后用make命令重新编译,还是会报同样的错误。抱歉,我是一个刚刚使用这个库的新学生,可能有一些不标准的东西。所以这就是我所说的重建版本的意思? 错误:16:09:17 [App] 构建日期:2023 年 3 月 9 日,16:07:01 16:09:17 [App] CPU:第 12 代 Intel(R) Core(TM) i7-12700KF(16 核)1​​6 :09:17 [App] RAM:38.55GB 物理内存 2.00GB 虚拟内存 16:09:17 [App] 操作系统:Linux 5.19.0-35-generic (x86_64) 16:09:17 [App] 磁盘:117.81GB (195.31GB) 空间 16:09:17 [App] 检测到 SSE 和 AVX 兼容的 CPU 和操作系统 16:09:17 [App] 命令行:DensifyPointCloud scene.mvs --resolution-level 1 16:09:17 [App]加载相机模型:平台0;相机 0;f 0.422x0.422;姿势 1162 16:09:17 [应用程序] 从界面格式加载的场景(109 毫秒):1162 张图像(1162 校准),总共 9191.60 MPixels(7.91 MPixels/图像)426067 个点,0 个顶点,0 个面 16:09:17 [App] 发现摄像头没有指向场景中心;场景将被视为无界(无 ROI)16:09:

  • 可见性信息(1477808 次观看 - 3.47 次观看/点): 0 次观看 1 次(0.00%) 254544 次观看 2 次(59.74%) 73345 次观看 3 次(17.21%) 98178 次观看 4次以上(23.04%) 2 分钟/3.46849 平均值(3.98363 标准差)/109 最大值 16:09:24 [App] 准备用于密集重建的图像已完成:1162 张图像(6s687ms) 16:09:25 [App] 选择用于密集重建的图像已完成:1162 张图像( 797ms)估计深度图 111 (9.55%, 11m6s, ETA 1h45m)...在抛出 'cv::Exception' what() 实例后调用终止:OpenCV(4.5.4) ./modules/imgproc/src/ resize.cpp:4054: 错误: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'

中止(核心转储)

我遇到了同样的问题,你解决了吗?

can you input chinese? I'm not sure what the problem is, but the photos I took with a different device didn't have the problem. It actually works pretty well. I speculate that some parameters of the photo should be the problem.

你计算深度图用的GPU还是CPU,可以给我共享一些你的数据吗?我正在测试这个问题

我用小米12Pro拍摄的照片都可以正常计算

之前出错的数据还有吗,上面的链接失效了

没有了,我的谷歌网盘不够就删掉了,上面那段数据是会出错, 对你应该没有任何帮助。

给一下你的邮箱,可以吗?这样效率比较低

I met the save problem,have you solved this problem?