When you take a picture in the multi-piece tagging experience, it is passed through our machine learning models that perform two tasks - Object Detection (to identify where the litter is in the photo) and Classification (to identify the object, material, and brand classification of said litter).

Once the image has been processed, based on whether or not the model has detected the object (litter) in the photo, it will associate a count to it (number of litter pieces in one photo). This number is then is added to the user’s profile under impact.

In the case that the model is unable to identify where the object is in the image, the image will still appear in the gallery but will not be counted towards your impact. You will have to manually review and add the object and tags.

To Manually Add Objects Follow These Steps:

  1. First, open your Gallery.

  2. Select the photo that has "0" pieces in it.

3. Once you've clicked on the image, it will show you that the LitterAI was unable to detect any litter in the picture. You can now add a litter count to the image manually. Click on the "+"

4. You will now be able to add the quantity per object. (in the photo above there is 1). You can also add the object, material, and brand tags, and "other" interesting tags of your choice.

5. You can also increase the quantity of this object (eg: If there is more than 1 tissue paper in the photo).
6. You have the ability to add another object too (eg: if the image consists of tissues and another type of litter, like cigarettes for instance).

7. Once you are done adding all the relevant objects and tags, simply click on "I've reviewed all tags"

8. Now the number will get updated in your gallery as well as under your Profile under your own Litter Map.

And That's It!

You should see a more accurate litter count reflected on your activity page. If you're still having issues, or need other support, please contact us at support@litterati.org, or send us a direct message on Facebook, Instagram, or Twitter.

Did this answer your question?