0

I am running a small experiment to crowd source images from users to build mesh models of interiors. I have been working with the ReCap API and it throws me a progress error when I try it with 160 images. My steps are with 40,80,120 and then 160 images. It worked alright for the first three sets.

Is this something to do with the engine itself? My fundamental procedure is the same. I have encountered similar errors before when the mesh can not be reconstructed properly by the ReCap engine and it does not throw back an error code. 

{
    "Usage": "0.64602208137512",
    "Resource": "/photoscene/nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg/progress",
    "Photoscene": {
        "photosceneid": "nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg",
        "progressmsg": "ERROR",
        "progress": "100"
    }
}

And here are the details of the processed photoscene:

{
    "Usage": "0.6743381023407",
    "Resource": "/photoscene/nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg/properties",
    "next_token": {},
    "Photoscenes": {
        "Photoscene": {
            "photosceneid": "nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg",
            "itemName": "nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg",
            "workerID": "3.0.0|10.64.128.165|D:\\stratos-worker\\src",
            "clientID": "BfKCfeWsfoeU8cnAz8qehKNbQzFZtD7y",
            "clientStatus": "SENT",
            "workerStatus": "ERROR",
            "type": "all",
            "userID": "IR9sKbyo5tJVcovRLItxGZBQ43LFGwC1R5r19WkE6wQ=",
            "nbShots": "000161",
            "sendEmail": "true",
            "inFilename": "3.0.0/IN/Mb9NMJTN5Pbeba5vJ4viOooSP4tPDNGFSa1gXe5jtoE=/nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg-IN.3dp",
            "convertStatus": "ERROR",
            "projectID": "nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg",
            "engineVersion": "3.0.0.3104",
            "convertFormat": "obj,rcm",
            "userAgent": "PF_APIv3.2.412-Photofly-WebAPI-FORGE",
            "KPStatus": "ERROR",
            "creationDate": "2019-05-08T15:21:13",
            "name": "P4_Unguided_01",
            "outFilename": "3.0.0/OUT/810ddyYgIdPmZE8VUph1DvDmHdY12FgknQaiks1VdV0=-nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg/100000000/P4_Unguided_01.3dp",
            "translationScope": "error",
            "Queue_messageID": "ca0a5998-be6f-435e-a5b1-6ad70cd5e6a6",
            "Queue_message": "STITCH nsyAf7qWc2QDwa1Brpq37YkXc3t8k1dNyrp5uYSGhCg 2.0 3.0.0.3104 0",
            "Queue_date": "2019-05-08T15:54:30+00:00",
            "maxResolutionForImage": "100000000",
            "Queue_name": "forge-bigOnAWSphotoflyqueue-calib",
            "UseTitanTextureEngine": "1",
            "workerType": "g3.8xlarge",
            "scenetype": "object",
            "processingtime": "1692.273",
            "progress": "100",
            "worker_version": "2.0",
            "aws_hourly_rate": "2.156000",
            "instanceID": "i-0eeac8cca7a8087ab",
            "version_worker": "0.5.0.248",
            "version_engine": "19.4.3.43",
            "worker_temp": "D:\\Temp",
            "purchase_option": "spot",
            "status": "ERROR",
            }

All the files are uploaded, not pasting all the files due to body sizing limit on code snippet.

Nimalan_Sri
  • 11
  • 2
  • When you are doing this steps 40, 80, 120 and so on, are you creating a separate scene for each one, or you are adding images and retriggering the process? – denis-grigor May 09 '19 at 20:09
  • One thing you could try is to use a different reconstruction engine version by specifying `version=1.0` in the `POST /photoscene` call. If that doesn't work could you send the photos to us so we can look into them (via a sharing service like weTransfer/Dropbox and send the download link to forge.help@autodesk.com)? – Bryan Huang May 10 '19 at 02:19
  • @denis-grigor I am indeed creating a seperate scene for each step. the purpose is to study separate scenes individually and cumulatively. – Nimalan_Sri May 14 '19 at 12:57

0 Answers0