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.

Topics - regiser

Pages: [1]
Feature Requests / Please support for Dual Camera Mapping System
« on: March 14, 2020, 08:50:56 AM »
Please provide support for Dual Aerial Camera Mapping System Data Processing,like PhaseOne's iXU-RS1900, especially after aligning photos successfully, then the program stitch the two photos from the same exposure station into one distortion-free、perspective large-format image for stereoplotter workstation.

Hi everyone, When I try to use Agisoft MetaShape 1.5.3 to process relatively large aerial dataset, at the stage of “Align Cameras”, after 75% completed,there are so many “merging failed” problems in the merging blocks step. So, at the final step, there will be many very small blocks only consist few photos(<30 Photos).To get the precise result, the program need many iterations to eliminate the small blocks. When large dataset, every iteration need very long time, so the overall processing time maybe unacceptable(>10 hours to finish the last 5%).So I want to know why this happened? And is there any solution to solve this problem.

Feature Requests / New Features Supporting Deep Learning
« on: March 19, 2019, 04:48:44 AM »
Very glad to hope Agisoft team in the near future will add new features supporting DL:
1.New features can make use of the state-of-the-art pre-trained model to identify objects or segment objects from the mosaic;
2.New features can use pre-trained network model as the basis to learn more user's local samples, in another word, that means users can train their own DL network module;     

Feature Requests / Multi-rig support for synchronous cameras
« on: May 11, 2017, 08:59:47 AM »
Now, I can bundle the synchronous cameras to a single camera station manually or through the Python script.However,when I have a project over 10000 images, the script should have many loops over every image on each camera station, so that is too time-consuing, very slow.Can the Dev team add a new function that automatically recognize the camera station through image name or positions.Or even add a new time delay parameter for each image in the bundle adjustment math model.

     Anybody knows the basic principle or algorithm behind the function of 'Estimate Image Quality'?

Pages: [1]