@Augmented_AI
Thank you so much for the Tutorial Series, I really benefited from it :D. Im getting 7 FPS on rpi, with a fast microSD card
@davidisrael7534
Hi, great video. is it possible to run R-CNN mask on RPI?
@not_a_number
Hi Mark！ Cool tutorial series, now that Movidius NCS II is released, I wonder how much the performance will be improved, might try and report back. BTW, say if I try to live counting how many people are in the frame but counting nothing else, which model would you recommend to get optimum performance?
@hashirali2604
Maybe I don't have to ask here but I need to know that is it possible to annotate different objects and then train them one by one for the same model, if yes then tell me how? Or I must have to annotate all the objects at once and then train that model... Also how can I label objects in label.txt file means it should be like "birdcatboat" or like "bird,cat,boat" or like "Bird, Cat, Boat" Is there any restrictions or something? Your answer will be much appreciated... Thank you
@nikhilwadibhasme9608
Thanks mark....awesome you are.can you please provide information in human pose estimation.i saw your video on pose estimation. We get the human body points (co ordinates). Can we use those coordinates to predict human pose (example - sitting , walking)
@mokera3801
can I train my own model and load it in NCS ? that would be great if you make a video about that. Many Thanks
@jrub3314
I love U! ♥️♥️♥️
@23jimi
How did you go with training a custom model for the NCS? In an earlier video you said you were having problems getting the fidget spinner model running. Any success?
@ShaanCoding
Brillant tutorial. Great work once again Mark! Hope Intel honours the warranty for the NCS. Curious what fps do you get in the modulus on 128*128? I'm saving up for a Movidius and am wondering how well it does on those resolutions.
@Kuuey
In your opinion do you think the technology is able to track a ball and count how many times it bounces? This technology amazes me, I have zero experience related to coding but this is amazing! Thank you for all the content, I hope to one day understand this!
@ffffffffffy
How would you train this Mobile SSD model to recognize a new object in Caffe?
@leonyoung7383
Thanks for the video. It help me a lot. And I wonder if there is some resources for teaching me how to train my model by tensorflow and use it by NCS. That would be helpful if you can make a video about that. And I am really benefited by this video. Thanks so much.
@10gerch
The test were made on the rock64 1GB or in the rock64 4GB?
@MrSpaceboyy
Use google Coral in future videos please
@danishbabjan
Mark, how do you make it detect a person only? I tried deleting all the other classes except "person' from CLASSES and it gave me an error like this:
Traceback (most recent call last):
File "webcam.py", line 45 in <module>
clss = CLASSES[int(output[i + 1])]
IndexError: string index out of range
Sidetrack: I want to make it detect a person only and when a person is detected, it will trigger something like a buzzer. All running from the Raspberry Pi.
@m.a.qurishee9220
Thanks for the video. Can you please give me a code for counting four objects such as 2 bottle, 1 monitor and 1 doll. I am struggling to find the exact code. Thanks
@ZoeyR86
Might be usb 2.0 bandwidth on the pi try it with a monitor on the pi and the network unplugged the ethernet is tied to the usb
@chuliut8173
HI, mark ! the ROCK64 +NCS How fast the frame? 10fps? or can use another device making it better?
@ffffffffffy
Sweeet
@robertparenton7470