Badly formatted text on logoAnnotations responses

48 views
Skip to first unread message

Luis Carneiro

unread,
Aug 1, 2016, 4:38:39 AM8/1/16
to cloud-vision-discuss
For some requests, the logoAnnotations text responses come with all words "glued".

Example, for this Picasso Guernica image:


One of the description text was: "Guernica,Photomosaicmountedonaluminum". Like this, it's impossible to separate the words and conveniently handle the text.

This does not happen for all images. I have seen several images where the description text is well formatted and readable.

Is there any chance of getting the text well formatted? Is this an issue you are working on?

Ram Ramanathan

unread,
Aug 1, 2016, 1:58:26 PM8/1/16
to Luis Carneiro, cloud-vision-discuss
Apologies for the error. Let us investigate this and get back to you


--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-dis...@googlegroups.com.
To post to this group, send email to cloud-visi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cloud-vision-discuss/74ee92ab-2da0-4a29-a7ad-58942a94ec62%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Luis Carneiro

unread,
Aug 3, 2016, 5:42:55 AM8/3/16
to cloud-vision-discuss, elt...@gmail.com
Thank you very much for your fast reply!

I'll stay stunned.


On Monday, 1 August 2016 18:58:26 UTC+1, Ram Ramanathan wrote:
Apologies for the error. Let us investigate this and get back to you


On Mon, Aug 1, 2016 at 1:38 AM Luis Carneiro <elt...@gmail.com> wrote:
For some requests, the logoAnnotations text responses come with all words "glued".

Example, for this Picasso Guernica image:


One of the description text was: "Guernica,Photomosaicmountedonaluminum". Like this, it's impossible to separate the words and conveniently handle the text.

This does not happen for all images. I have seen several images where the description text is well formatted and readable.

Is there any chance of getting the text well formatted? Is this an issue you are working on?

--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-discuss+unsub...@googlegroups.com.

Luis Carneiro

unread,
Aug 19, 2016, 5:58:51 AM8/19/16
to cloud-vision-discuss, elt...@gmail.com
Hi Ram,

Any news regarding this issue?

Thanks in advance,
Luis Carneiro


On Monday, 1 August 2016 18:58:26 UTC+1, Ram Ramanathan wrote:
Apologies for the error. Let us investigate this and get back to you


On Mon, Aug 1, 2016 at 1:38 AM Luis Carneiro <elt...@gmail.com> wrote:
For some requests, the logoAnnotations text responses come with all words "glued".

Example, for this Picasso Guernica image:


One of the description text was: "Guernica,Photomosaicmountedonaluminum". Like this, it's impossible to separate the words and conveniently handle the text.

This does not happen for all images. I have seen several images where the description text is well formatted and readable.

Is there any chance of getting the text well formatted? Is this an issue you are working on?

--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-discuss+unsub...@googlegroups.com.

Luis Carneiro

unread,
Nov 2, 2016, 12:24:57 PM11/2/16
to cloud-vision-discuss
Hi Ram,

We still have this issue. Can we have any update on this please?

Regards,
Luis

Luis Carneiro

unread,
Jan 4, 2017, 12:07:02 PM1/4/17
to cloud-vision-discuss
Hi Ram,

Do we have any update regarding this issue?

We are close to release our application and this is an issue we don't want our users to see. If we don't get any update about a fix on this we are sorry but will need to leave your service and start using an alternative.

Regards,
Luis

Ram Ramanathan

unread,
Jan 4, 2017, 12:08:20 PM1/4/17
to Luis Carneiro, cloud-vision-discuss
Luis - Let me follow up within couple of days

--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-dis...@googlegroups.com.

To post to this group, send email to cloud-visi...@googlegroups.com.

Luis Carneiro

unread,
Jan 4, 2017, 12:10:58 PM1/4/17
to cloud-vision-discuss, elt...@gmail.com
Tks Ram. We have been asking for a solution for this since August, therefore we look forward to hear an update regarding this.


On Wednesday, 4 January 2017 17:08:20 UTC, Ram Ramanathan wrote:
Luis - Let me follow up within couple of days

On Wed, Jan 4, 2017 at 9:07 AM Luis Carneiro <elt...@gmail.com> wrote:
Hi Ram,

Do we have any update regarding this issue?

We are close to release our application and this is an issue we don't want our users to see. If we don't get any update about a fix on this we are sorry but will need to leave your service and start using an alternative.

Regards,
Luis

On Wednesday, 2 November 2016 16:24:57 UTC, Luis Carneiro wrote:
Hi Ram,

We still have this issue. Can we have any update on this please?

Regards,
Luis

On Monday, 1 August 2016 09:38:39 UTC+1, Luis Carneiro wrote:
For some requests, the logoAnnotations text responses come with all words "glued".

Example, for this Picasso Guernica image:


One of the description text was: "Guernica,Photomosaicmountedonaluminum". Like this, it's impossible to separate the words and conveniently handle the text.

This does not happen for all images. I have seen several images where the description text is well formatted and readable.

Is there any chance of getting the text well formatted? Is this an issue you are working on?

--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-discuss+unsub...@googlegroups.com.

Ram Ramanathan

unread,
Jan 4, 2017, 6:45:34 PM1/4/17
to Luis Carneiro, cloud-vision-discuss
Are you still seeing this issue - I ran through the exact same image at https://cloud.google.com/vision
here is the response
{
  "landmarkAnnotations": [
    {
      "mid": "/m/039_n",
      "description": "Guernica to Gernika",
      "score": 95,
      "boundingPoly": {
        "vertices": [
          {
            "x": 45,
            "y": 31
          },
          {
            "x": 739,
            "y": 31
          },
          {
            "x": 739,
            "y": 279
          },
          {
            "x": 45,
            "y": 279
          }
        ]
      },
      "locations": [
        {
          "latLng": {
            "latitude": 43.315733,
            "longitude": -2.6799999999999997
          }
        }
      ],
      "lat": 43.315733,
      "long": -2.6799999999999997,
      "latText": "43.315733",
      "longText": "-2.680000"
    },
    {
      "mid": "/m/01jcyh",
      "description": "Museo Nacional Centro de Arte Reina Sofía",
      "score": 94,
      "boundingPoly": {
        "vertices": [
          {
            "x": 18,
            "y": 28
          },
          {
            "x": 762,
            "y": 28
          },
          {
            "x": 762,
            "y": 279
          },
          {
            "x": 18,
            "y": 279
          }
        ]
      },
      "locations": [
        {
          "latLng": {
            "latitude": 40.408267,
            "longitude": -3.694217
          }
        }
      ],
      "lat": 40.408267,
      "long": -3.694217,
      "latText": "40.408267",
      "longText": "-3.694217"
    }
  ],
  "logoAnnotations": [
    {
      "mid": "/m/0421l5",
      "description": "Guernica",
      "score": 92,
      "boundingPoly": {
        "vertices": [
          {
            "x": 14,
            "y": 12
          },
          {
            "x": 783,
            "y": 12
          },
          {
            "x": 783,
            "y": 274
          },
          {
            "x": 14,
            "y": 274
          }
        ]
      }
    },
    {
      "description": "Guernica, c.1937",
      "score": 89,
      "boundingPoly": {
        "vertices": [
          {
            "x": 18,
            "y": 22
          },
          {
            "x": 756,
            "y": 22
          },
          {
            "x": 756,
            "y": 274
          },
          {
            "x": 18,
            "y": 274
          }
        ]
      }
    },
    {
      "description": "Guernica, Photomosaic mounted on aluminum",
      "score": 74,
      "boundingPoly": {
        "vertices": [
          {
            "x": 19,
            "y": 31
          },
          {
            "x": 770,
            "y": 31
          },
          {
            "x": 770,
            "y": 280
          },
          {
            "x": 19,
            "y": 280
          }
        ]
      }
    }
  ],

On Wed, Jan 4, 2017 at 9:10 AM Luis Carneiro <elt...@gmail.com> wrote:
Tks Ram. We have been asking for a solution for this since August, therefore we look forward to hear an update regarding this.


On Wednesday, 4 January 2017 17:08:20 UTC, Ram Ramanathan wrote:
Luis - Let me follow up within couple of days

On Wed, Jan 4, 2017 at 9:07 AM Luis Carneiro <elt...@gmail.com> wrote:
Hi Ram,

Do we have any update regarding this issue?

We are close to release our application and this is an issue we don't want our users to see. If we don't get any update about a fix on this we are sorry but will need to leave your service and start using an alternative.

Regards,
Luis

On Wednesday, 2 November 2016 16:24:57 UTC, Luis Carneiro wrote:
Hi Ram,

We still have this issue. Can we have any update on this please?

Regards,
Luis

On Monday, 1 August 2016 09:38:39 UTC+1, Luis Carneiro wrote:
For some requests, the logoAnnotations text responses come with all words "glued".

Example, for this Picasso Guernica image:


One of the description text was: "Guernica,Photomosaicmountedonaluminum". Like this, it's impossible to separate the words and conveniently handle the text.

This does not happen for all images. I have seen several images where the description text is well formatted and readable.

Is there any chance of getting the text well formatted? Is this an issue you are working on?

--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-dis...@googlegroups.com.
--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-dis...@googlegroups.com.

To post to this group, send email to cloud-visi...@googlegroups.com.

Luis Carneiro

unread,
Jan 5, 2017, 5:27:49 AM1/5/17
to cloud-vision-discuss, elt...@gmail.com
Hi Ram,

Ok, I re-run the request on this image and yes, the words are not glued. The issue was related with our code indeed. We were applying a regex to remove all the clutter (tabs, newlines and so on) from the json response so we could store in on our database for future reference. It happens that regex was causing this issues of glued words.

Thanks for your time and apologies for the confusion.

Regards,
Luis
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-discuss+unsub...@googlegroups.com.
To post to this group, send email to cloud-visi...@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "cloud-vision-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cloud-vision-discuss+unsub...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages