Blind and Visually Impaired Community
↑
4
↓
Second attempt (self.Blind)
submitted by DNNenthusiast
I posted a link to the survey about a month ago but received 0 replies.
I send the link to some organizations but it seems like they are not forwarding it to the end users.
It is really hard to collect data during the lock down and online surveys are the only option for me.
So, here we go again.
I am a second-year PhD student and my main research focus is the development of technology to support visually impaired people. For the last six months, I was working on an app that will help VIP to explore tactile images independently and allow them to gain clarifying audio feedback. I trained a deep neural network to detect the user's fingertip and the results were surprisingly good. The model works pretty smooth even at poor lighting conditions and can detect each finger independently. Now I am trying to get some insights to improve the overall user experience and interface design.
I looked at the previous posts and noticed that some users got annoyed with the endless survey requests. I would like to apologize for taking your time in advance but if you decide to participate, it will take you only 15 minutes to complete. This work is not a high school or master's degree project (with all respect to their work). In total, four years of research will be dedicated and we are planning to produce a meaningful and functional product that will really help people. Also, in the end, we will share all models and algorithms with others by making it open-source. Here is the survey $1 https://forms.gle/MeNgJDic8eiugG6g9
Thanks for your time!
I send the link to some organizations but it seems like they are not forwarding it to the end users.
It is really hard to collect data during the lock down and online surveys are the only option for me.
So, here we go again.
I am a second-year PhD student and my main research focus is the development of technology to support visually impaired people. For the last six months, I was working on an app that will help VIP to explore tactile images independently and allow them to gain clarifying audio feedback. I trained a deep neural network to detect the user's fingertip and the results were surprisingly good. The model works pretty smooth even at poor lighting conditions and can detect each finger independently. Now I am trying to get some insights to improve the overall user experience and interface design.
I looked at the previous posts and noticed that some users got annoyed with the endless survey requests. I would like to apologize for taking your time in advance but if you decide to participate, it will take you only 15 minutes to complete. This work is not a high school or master's degree project (with all respect to their work). In total, four years of research will be dedicated and we are planning to produce a meaningful and functional product that will really help people. Also, in the end, we will share all models and algorithms with others by making it open-source. Here is the survey $1 https://forms.gle/MeNgJDic8eiugG6g9
Thanks for your time!
rp-turtle 3 points 3y ago
I have a few questions.
1. What exactly does your technology do? As I drag my finger tips over an image, it describes what’s under my fingers? Like seeing AI already does?
2. Is this for digital images like on a phone or will the image need to be printed and put on some sort of device that identifies where my fingers are by reading them through the image? I can only see that being useful in like a Museum or art display.
3. Is the app something you would point at an image I am feeling with my hands and it could describe the scene from just using the camera to identify where my fingers are touching on the image?
1. What exactly does your technology do? As I drag my finger tips over an image, it describes what’s under my fingers? Like seeing AI already does?
2. Is this for digital images like on a phone or will the image need to be printed and put on some sort of device that identifies where my fingers are by reading them through the image? I can only see that being useful in like a Museum or art display.
3. Is the app something you would point at an image I am feeling with my hands and it could describe the scene from just using the camera to identify where my fingers are touching on the image?
DNNenthusiast [OP] 2 points 3y ago
1. Yes, predefined audio information is triggered as the user drags his/her finger. It is different from Seeing AI in a way that everything is processed on the device (not cloud based). It needs an internet connection only to download a small JSON file with the information about an image. This is easily done by scanning a QR code placed on the image. Also, it will be available on Android.
2. Tactile image has to be printed on a swell paper. After you have a printed image no additional devices except your phone are needed. Also, I am developing an online service with the ready-to-print images available and where users can share their own designs. Yes, the system can be used in a museums but I am developing it for educational purposes.
3. Yes. Also, the app is capable of detecting different gestures. So, the user can request an additional info, given that this information is provided.
Hope I was able to answer your questions))
2. Tactile image has to be printed on a swell paper. After you have a printed image no additional devices except your phone are needed. Also, I am developing an online service with the ready-to-print images available and where users can share their own designs. Yes, the system can be used in a museums but I am developing it for educational purposes.
3. Yes. Also, the app is capable of detecting different gestures. So, the user can request an additional info, given that this information is provided.
Hope I was able to answer your questions))
rp-turtle 1 points 3y ago
Yes, you answered my questions. I participated. Good luck!
DNNenthusiast [OP] 1 points 3y ago
Thanks!!!
bjayernaeiy 1 points 3y ago
I'll give it a look
Superfreq2 1 points 3y ago
Don't take it personally about the organizations not responding thing. They are terrible for that even at the best of times unless you approach them as an established business.
I know that some of that comes from not wanting to expose their members to untested or badly formulated concepts, as they would be less likely to take emails from the organization seriously if that happened allot, but IMO they are also just badly staffed.
I know that some of that comes from not wanting to expose their members to untested or badly formulated concepts, as they would be less likely to take emails from the organization seriously if that happened allot, but IMO they are also just badly staffed.
DNNenthusiast [OP] 1 points 3y ago
Thanks for the information, it makes sense now. I have sent emails almost to 50 organizations across the UK but got no feedback. So, online forums is my only option now.
bradley22 0 points 3y ago
I wouldn’t find this useful but you could try posting to the RNIB.
DNNenthusiast [OP] 1 points 3y ago
Thanks!
bradley22 1 points 3y ago
No problem.
This nonprofit website is run by volunteers.
Please contribute if you can. Thank you!
Our mission is to provide everyone with access to large-
scale community websites for the good of humanity.
Without ads, without tracking, without greed.
scale community websites for the good of humanity.
Without ads, without tracking, without greed.
©2023 HumbleCat Inc • HumbleCat is a 501(c)3 nonprofit based in Michigan, USA.