a error "Image is too small" occurs when detect face

[qinyuanpei]

OK, there is a error "Image is too small" occurs when I try to detect face in image to upload
Here are results returned by API:

{
    u'error': 
    {
        u'message': u'Image size is too small.', 
        u'code': u'InvalidImageSize'
    }
}

Actually, all width and height of these images are more than 640px.
These images can be handled when I upload them into your website, so I am so confused about this result. I write Python script as following:

#!/usr/bin/python
#-*- coding: utf-8 -*-

import json
import requests

class FaceAPI:

    def __init__(self,secretKey):
        self.baseUrl = 'https://westcentralus.api.cognitive.microsoft.com'
        self.secretKey = secretKey

    def detectFace(self,image):
        # Request headers.
        headers = {
            'Content-Type': 'application/octet-stream',
            'Ocp-Apim-Subscription-Key': self.secretKey,
        }

        # Request parameters.
        params = {
            'returnFaceId': 'true',
            'returnFaceLandmarks': 'false',
            'returnFaceAttributes': 'age,gender,headPose,smile,facialHair,glasses,emotion,hair,makeup,occlusion,accessories,blur,exposure,noise',
        }

        # Body. The URL of a JPEG image to analyze.
        imgData = open(image).read()

        # Call REST API
        try:
            response = requests.post(
                self.baseUrl + '/face/v1.0/detect',
                json = None,
                data = imgData,
                headers = headers,
                params = params
            )

            if(response.status_code == '200'):
                data = json.loads(response.text)
                return (True,data)
            else:
                data = json.loads(response.text)
                return (False,data)
        except Exception as e:
            return (False, e.message)

So can you help me to resolve this issue? Thanks you.

[fatahfattah]

Hi qinyuanpei, I'm facing the same issue right now.
Did you manage to find a fix?

[qinyuanpei]

I submit this issue at last year, but I can't solve it yet. 😃

[fatahfattah]

Ah okay, i'll just try a manual api call. thanks

[chetanyachopra22]

I am facing the same issue is there any fix to this problem yet
does manual call worked ?? @fatahfattah

[fatahfattah]

@chetanyachopra Yes manual implementations work, can't recall what the issue was. I'd suggest to do all the calls manually though, not just this one.

[fadyanwar]

Just wondering, if there have been any updates for this issue.

[Soorya018]

@chetanyachopra @fatahfattah @qinyuanpei @fadyanwar Is there any ways of resolving this issue?

[sabrimev]

I see there is no solution here yet but want to share how we fix this on our side:

We used to detect the face and then creating a person within a group right after detection. For these both operations we were using the same MemoryStream object so when I check the MemoryStream object right before the second operation I saw that the position of the stream was the end of the size. I just set the position to 0 again so everything started to work fine.

Be cautious when we try to use the same MemoryStream object multiple times. At first usage, the position is 0 but for the next usages the position will be the end file and you need to set it to 0 again to be able to use.