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.

Messages - Phogi

Pages: 1 2 [3] 4 5 ... 7
Python and Java API / Help in localframe explain
« on: November 22, 2019, 06:48:04 AM »
Hi everyone,

Could anyone give some hints about the matrix?
Hello claranine,

Code: [Select]
sourceGeogr = camera.reference.location #source values in geographic coordinate system
estGeogr = #estimated position in geographic coordinate system

sourceGeoc = #measured values in geocentric coordinates
estimGeoc = chunk.transform.matrix.mulp( #estimated coordinates in geocentric coordinates
local = #local LSE coordinates
error = local.mulv(estim - source)

As in this post why the error needs to use local.mulv to transform rather than directly use (estim - source)? Is this due to unit difference?

Thanks in advance!

Bug Reports / Image quality estimation not right for special blurry case
« on: November 19, 2019, 04:20:18 AM »
Hi Alexey,

I had one example of image with "blurry" due to out of focus, but the image quality estimator still gives me 0.76 image quality which says it is good. Is there any function to detect this kind of blurry?

Thanks a lot!

General / Re: Agisoft Metashape 1.6.0 pre-release
« on: November 14, 2019, 07:19:59 AM »
The new release looks so good, with several new settings in alignment and empower the processing.

I got two questions about the new release:
(1) I find that camera calibration part set cx,cy as one tick together, and then omitted the p3 p4 parameters, is there any relationship to the new fit additional adjust function? If that is adding p3p4 in, would this add extra error?

(2) if the estimated alignment function should be used after using "source" method and then do not "reset alignment"?

Thanks for answering this!

General / Re: Export Tiled Model slpk format Local coordinate
« on: August 15, 2019, 03:10:37 AM »
Hi Alexey,

Is the SLPK file projection fixed? I tried an export it showed me EGM2008 in v1.5.2 build 7838.


General / Re: Is it possible to remove mask after imported it?
« on: July 23, 2019, 09:48:30 AM »
Thank you! Is it able to reset from API?

General / Is it possible to remove mask after imported it?
« on: July 19, 2019, 05:22:37 AM »
Hi friends,

Is there a button to remove imported mask? I haven't found it... Could anyone help? Or it can only be replaced by new mask file?


Python and Java API / Python script modify camera Yaw Pitch angles
« on: November 26, 2018, 10:19:16 AM »
Dear Support,

I had some searches in the forum but haven't found similar answers, so I post this new topic here.

What I want to achieve is, when I loaded the photos with orientation angles, is it possible to use python script to modify the angles rather than load a modified geotag file separately from GUI? For example an image Y/P/R is 17.2 / 0.5/0, and I want to change it from python to 180/89.5/0, is it possible to do from python api?

Thanks a lot!

General / Re: Agisoft Metashape 1.5.0 pre-release
« on: November 19, 2018, 09:52:45 AM »
Hi Alexey,

Thanks for the great work, I just tried the auto-classification function, and I have two questions to report:

(1) After the autoclassification, I can't reset the classification. However if I continue used "Classify Ground Points" the point clouds will be classified based on the "ground point" classes.

(2)  Some ground in "Classify points" were clasified into building, since it is based on machine learning, would it be possible to adjust any parameters or it is all default set? Also the time is much longer than previous tool, I got 52.7595 min vs 6.750816667 mins.

Kind regards

Dear Alexey,

May I know what's the unit of marker.position? I searched some threads and know it's in camera plane, is it what is used for (X, Y, Z) - point coordinates in the local camera coordinate system? For a pixel in the camera, could you give an example on how it gets the position values? Also the u/v in camera calibration value would it be expected as millions or few handreds if it's pixels?

My another question is about the classification function, the angle/distance settings are clear, but could you offer some articles on how they are working? Cause sometimes I filter a conveyor and it removes the other points for different densities. Also if the automatic classification can add other classes other than ground?

Thanks a lot!

Hi Alexey,

Ah sorry I found the geoid file I downloaded are having a problem, redownloaded and it works. Also I found it works under NAD83 (2011) but not on NAD83. Just a possible bug report but generally it works now.


Hi Alexey,

I'm downloading the new 1.4.4 version and saw the same problem, I put the geid file under /ProtoScan Pro/geoids folder. Is it changed to some new route?


Python and Java API / Re: select cameras by time
« on: August 21, 2018, 06:07:34 AM »
Hi Alexey,

Thank you! That script helps hugely! I now can split by time, however I'm still not proficient to split and move them into different chunk, or automatically split dictionary into chunks. However I can manually right click and move to new chunk. Just noticed one thing is when I move to new chunk the other photos just moved in chunk2, if I don't click in the reference panel to cancel selected images (when I'm in chunk 1), the images in chunk 2 will be moved together with the ones selected in chunk 1 to the new chunk.

My crappy codes are:

Code: [Select]
import PhotoScan, time

STEP = 60 #seconds

chunk =

cameras = dict()
for camera in chunk.cameras:
    cameras[time.mktime(time.strptime(['Exif/DateTime'], '%Y:%m:%d %H:%M:%S'))] = camera
#    print('camera',camera)
#    print(cameras[time.mktime(time.strptime(['Exif/DateTime'], '%Y:%m:%d %H:%M:%S'))])
#print('Camera', cameras)

chunkcamera = chunk.cameras

sorted_by_time = list(cameras.keys())

chunk_1 = dict()
chunk_2 = dict()
chunk_3 = dict()
chunk_4 = dict()
chunk_5 = dict()
tmpdict = dict()
chunk_initial = 1

for i in range(0, len(cameras)-1):
    tmpkey = []
    tmpkey = sorted_by_time[i]
    if sorted_by_time[i] - sorted_by_time[i-1] < STEP and chunk_initial < 6:
        if chunk_initial == 1:
#            print(cameras[1])
#            chunk_1.update((k,cameras[k]) for k in tmpkey)
            cameras[sorted_by_time[i]].selected = True
            cameras[sorted_by_time[i]].selected = False
    elif sorted_by_time[i] - sorted_by_time[i-1] > STEP and chunk_initial < 6:
        chunk_initial += 1
        print('More than 5 chunks, needs increase chunks')

Thanks a lot!

Python and Java API / select cameras by time
« on: August 16, 2018, 07:11:37 AM »
Hi Alexey,

Could you help me with sort camera by time in the python api? I tried

chunk =
cameras = chunk.cameras
cam_num = len(cameras)
chunk_N = 0
for i in range(0,cam_num-1,1):
    timestamp = time.mktime(time.strptime(['Exif/DateTime'], '%Y:%m:%d %H:%M:%S'))
    timestamp_next = time.mktime(time.strptime(cameras[i+1].photo.meta['Exif/DateTime'], '%Y:%m:%d %H:%M:%S'))
    if (timestamp_next - timestamp) <= 60:
        cameras.selected = True
        cameras[i+1].selected = True
        cameras[i+1].selected = False

But it is always not select the cameras, and sometimes select wrong cameras depending on the camera names. I am assuming it needs to sort by time first, however getting a list of camera time does not store the camera id correspondingly.

Thanks a lot for your help,

Kind regards,

General / Adaptive Camera Model Fitting which step
« on: April 11, 2018, 03:49:29 AM »
Hi Alexey,

In v1.4, the Adaptive camera model fitting option appears in both "Align Photos" and in "optimise cameras". Is there any difference? As in some threads it is mentioned rolling shutter should be in the sensor so it was moved out from optimising options, but what is the "Adaptive camera model" fitting option does now?

Thanks for your explanations and support!


General / Re: Camera FC6310 model problem
« on: April 05, 2018, 09:48:23 AM »
Hi Alexey,

I send to, do you get the report? Thank you very much!

Pages: 1 2 [3] 4 5 ... 7