1. For example: it detected a doll as cake, detected my book as pizza, detected my dog as a lion πππ, detected my sister as a doll π€£π€£, detected my bag as a balloon, detected my bike as trolly cycle.... But instead of it detected some right things like my bed, cycle, TV etc.
2. The AI needs more training and I'd be very happy to give it more examples of there were a training programs, as it wouldn't be the first deep learning training program in my history.
3. But its cool, just needs improvement.Model need more traning, In my experience I scan door and it's labelling it as Refrigerator.
4. Hoping for more accuracyAwesome idea but it needs a larger index as a lot of things came up as a toilet lolSir please add a feature.
5. Needs more training and would be great if transfer learning can be done in the future.
6. Collected data share to Bluetooth hc05.I need something similar to this but able to learn and able to speak what it sees?Not good.
7. Excellent advancement to what a mobile phone can doThe idea is awesome for non-conspirationist-boomers thinking that this app is spying on them, but they don't even know what the "ML" thing means.
8. But still needs improvement!! ππππHey, it could be a great help if you can give link of the project.
9. I am working on a problem and collaboration with this app seems perfectIts cool but sometimes gets things wrong.
10. ππThe app is very good and awesome but it detect something sometimes wrong.
11. false detection rate is high, cant detect objects far away and through pictures.