Zxen
Australia
Asked
— Edited
I need to extend the camera from it's circuit board and mount it in something like a cable dot on the side of the head facing forward. At the moment she looks borg. Not good. This is for department stores for women's fashion. Can I separate the camera with wires so its just a tiny dot?
How do I do that please?
PS The camera will definitely not be placed in the face of this particular mannequin head - It will be under the chin or facing forward from the ear.
Sorry just read your post again...not sure if that can be done but then again it could be.
From memory not sure if their is a ribbon, if there is then a ribbon extension cable would suffice.
If its too hard, can I buy a different camera that is compatible with the ezbv4 that I can control from ARC as normal?
I don't see why the camera de-cased wouldn't fit in there, you would be left with the exact same look in the end.
It has a really good heat sink on it and will be facing towards the back of course so there's no chance of it heating up plastics, etc.
Im currently finishing off the head on my project, and the de-cased camera fits in there easily, its quite a small head so you'll be fine.
There are other camera's that are compatible for sure.
I found this one - the first one I looked at: https://thepihut.com/products/adafruit-spy-camera-for-the-raspberry-pi
thank you i should have been more specific there.
Have you looking into the EZB camera with no shell that is in the developer's kit?
www.ez-robot.com/Tutorials/Lesson/34?courseId=4
Also, it looks like that same camera is used inside the JD robot head.
Regards,
Frank
ALAN built by Will Huff has the camera mounted in the upper head looking out thru a hole in the center of the forehead. I have ANTONN with the camera looking out the nose. Others have mounted cameras by the ears. The image can be rotated if you need to mount the camera other than it's normal way. Extension camera cables are available or can be built. DJ or someone from EZ Robot could tell you the maximum length allowed. What is the final look you desire ?
Edit:
You can mount it in the chin looking out a small hole. If tracking is to be used, see if relative tracking will do the job instead of grid tracking and maybe mount the camera in the neck or chest where it will be able to see and do what you want.
https://www.ez-robot.com/Community/Forum/Thread?threadId=4979 is the thread.
**Edit
DJ said in the thread about 8 feet of cable and the camera was working well.
Zxen, could you give more information to allow possible options?
At any rate, you can mount the camera on or in the head so it moves at the same rate as the head or on or in the torso so its stationary when the head moves. I've been told that you get better tracking if the camera is stationary and doesn't move when following an object.
I suggested the same in my first reply, but its essential that the camera be in a certain location, and the head is no longer an option.
Tricky scenario!
I'm suggesting that the camera be taken out of it's case and placed inside the robot so it's completely out of sight except for the small dark lens peeking through a small hole. Reading the OP's remarks it sounds like he want's the camera moved to a hidden spot and not hanging off the side of the head like a huge wart. Understandable. I don't think the camera lens needs to be removed. just place the small board with the lens still attached inside the robot.
If it had black eyes, then the camera could be fitted inside one of them, or even sunglasses, a neck chain with a pendant, etc.
But it seems the face needs to be clean with no signs of anything just a blank face.
Worse case scenario would be to not even install the camera on the mannequin, but rather behind it somewhere either hidden on a wall or on a stand, etc.
I know I could replace the white blank face with a different face that has eyes but that would be a different and much easier thing to achieve. I need to solve this one first. Blank face that can do facial tracking.
Can the camera be extended out of the board?
Can a different camera be plugged into the board?
Can a different board be used and the wires made compatible (easily) with ezbv4?
A small hole on the chin angled upwards to align with the mannequin's 'eye direction' is the most realistic suggestion I've heard so far, but I would rather not drill a hole in the chin - also this would probably require a lens over a big hole to counteract depth constraints. I wonder if a lens covering over a hole drilled into her forehead would look as ugly as a black dot - probably. The face looks good sheen.
Please keep the ideas coming - preferably about the camera board modifications and compatibilities. I'm quite convinced the camera is going to need to remain over her 'ear', but much smaller. If necessary I will stick half the naked board out of the side of her head so she has one ear(!) but I'd rather it was small and round.
Also, if you know any one way white material that has been invented (like one way mirror) than that would solve it. I guess if her face was gold or silver or black this would be a lot easier.
Maybe a 5mm periscope...
Sorry if i come across a little rude here but the chin would be taking a huge leap backwards...
Sorry i'm opting out, your asking the same questions over again which have all been answered.
Bye! Thanks for you're contributions! Alright everyone, we have to stop thinking now. All questions have been answered apparently.
sorry by opting out i mean i can't help any further, i just don't know enough about it sorry again.
1) The camera is part of the board so no...
2) already answered this in post#9
3) Again no to maybe possible but would be extremely difficult
It's not ridiculously difficult at all. Your camera control in ARC already supports this function. Click on the gear icon in the camera control. Near the top you'll see a few check boxes. Make sure Enable servo Tracking is checked and specify the servo that will move the head to follow the object. Then right below that box is one that says Track By Relative Position. When that is enabled the control assumes you have a stationary camera and moves your servo into position relative to the object being tracked. Just like magic!
Now you need to decide what needs to be tracked and set up and calibrate your camera accordingly. Of course you'll have to do this no matter which method you choose to use.
If you mount your camera in a stationary position I would not mount it on the wall away from the robot. The object will be tracked relative to where the camera is mounted and the head may not follow the object as well as needed. Try to place it on the torso as close to center as possible but if you mount it off center it should track just as well.
Your idea about mounting the camera lens under a material like a colored two way mirror is very interesting. That should work as long as the camera can clearly see the tracked object from all angles.
It needs to be stated that the quality of the image you will see in the camera control is not as important as you would think. The computer is not seeing the image like you and I do. Light and definition seems to be more important but I'm moving out into the weeds now and know very little about this. *eek* Perhaps others that have used this control more the me can add to why this is.
Newagetomy, sorry for fighting.
Richard, thanks for the clarity on board mods.
Dave, that would be amazing to have a stationary camera on a pole or wall, but my design most likely won't have one (she's going to be supported by the back of the neck and wear clothes). I still would be interested in what can be done if it results in a smoother motion. I assume a fish eye lens is required to see everywhere the head will look, though. 'It needs to be stated that the quality of the image you will see in the camera control is not as important as you would think.' If this is true then maybe a bit of white cotton material over the camera in her face would still allow facial tracking to work...
Here is what I have done to reduce the tumour. I was worried it would look like she had one ear, but I think it looks okay. If she was black or had eyes I would have put the camera in her face. Thanks again to Andy for telling me about screen rotation.
What do you guys think?
I will be interesting to see if your tracking will work as expected having the camera mounted on the side of the head like that. The camera won't have a full view of anything on the other side coming at it until the object passes the nose.
If you decide to keep the camera mounted like this perhaps you can put a wig on it and hide the camera in the hair.
If you're finding the movement to be jerky it because your camera is mounted on a swiveling base. The software is fighting to compensate for not only the movement of the object but also the swing of the camera. You may be able to smooth some of this out by adjusting your grid lines and some other fine tune operations in the settings area of the camera control. If you want a better shot of having smoother motion then my suggestion still stands; place the camera stationary in the torso and use the Relative Position Tracking option.
Good luck and please post a video of your final results. I'm fully out of suggestions. *sleep*
I plan on mounting and hiding my facial recognition camera behind Antonn's Bow Tie.
A centrally mounted camera would be best as Dave suggested, no additional lenses should be used.
After you do some camera trials for the tracking I think you will see the direction many of us have been leading you. Maybe try the remote stationary camera. Also realize a web / usb camera can also be used in the ARC software.
This would be connected directly to a computer which allows you to mount the camera anywhere. Also multiple cameras. A lot of options available.
'Also multiple cameras.'
Really?
That would be great. I can't think of how to apply it but I would love a link to a lesson or a project using interesting camera setups. I'd love the head being able to look people up and down and the robot says something like, 'Oh wow! Like, who are you wearing?' I might even put a microphone in her hand so she's like a red carpet journalist. Imagine if she could call people over based on colour recognition. 'Excuse me, you in the green. Come over here, I want to talk to you!' That would be considered AI. I also want her to mimic body movements using a Kinect sensor, which I consider an advanced mod - afterwards. I also want to give her a Pandorabots.com library of responses - I'll use an actress to voiceover her speech rather than rely on a computer voice.
People here in the Community can give suggestions and advice, but only you know what your needs are. Learn the basics and you will begin to gain a better understanding of what the EZB controllers can do with a camera.
By the way, the Palette robot seems to have a dot in the upper neck which could be a camera.
There is a lot of cool stuff available and things you can do with an EZ-Robot control. You have to look at example in the tutorials, and on the community forum.
Thanks for the Omron camera reference (http://www.ia.omron.com/products/category/sensors/vision-sensors_machine-vision-systems/smart-camera/index.html). I would have thought that was not compatible with the ezbv4 and such capabilities were driven by software. Having said that, I did speak about using a Kinect sensor, which has been unnecessarily difficult in my brief experiments (I guess I spent around 100 hours or so on it).
I found my other ez robot camera, and it's internals are different. I hope the new camera is the same design as the one I put in the side of her face, rather than this double decker one:
Try playing with 2 stationary cameras. See if you can track center to left with one and center to right with the other. This way you may scan left to right until someone comes by. Just a thought.
Edit:: try a web cam.
Ok. I will let you go organize your options and play with the EZB and cameras.
Incremental progress like that will prevent CPA as you put it.
If you are having difficulty mounting inside of a curved Head I could suggest using the new version 2 EZ-Robot camera and peel the lens away from the PCB to change the angle a little. It is just held against the PCB with double-sided tape.
My second suggestion would be to make only a 1mm hole somewhere in the mannequin's head. A 1mm hole is all you need for the actual lens portion to see.
Now you may have to drill out a small cavity for the rest of the lens but the opening only need to be 1mm. Here's an example, created with my amazing MS paint skills LOL.
I should also note that the camera v2 lens is interchangeable with Fish eyes lenses that are somewhat difficult to find but do exist. But note that you will need a larger viewing hole to incorporate a larger lens; it have would be 10mm in diameter.
Oh and yes I can confirm that you can definitely extend the camera cable, we've gone as far as 10 feet successfully but we've never really tested out the maximum range.
I have spent hours testing the facial and colour tracking and its just like I remember. EXTREMELY SLOW AND UNRELIABLE REACTION TIME.
The motors in my robot's head are fast - she can nod and shake her head faster than a real human. But when she is tracking, she moves like a pigeon. I get best results when I shrink the grid and I am moving my face further away, but I basically need to move past at less than 1km/h so she doesn't lose me.
Can someone please tell me how to fix this so its smooth?
Is it slow because it requires WiFi? Video is being sent wirelessly to my computer fast enough, so I don't understand whats causing this delay thats so extreme she moves in slow motion. Are the calculations being done in the ARC software or on the board? She only moves smoothly if I tell her to move 5 or more steps at a time, but this often results in overshooting the mark, after which her head begins to thrash violently to get back to the mark.
@ Zxen I feel your pain. I've tried to get smooth motion from tracking and the best thing to play with is the increments steps found in the camera set up where your servos are set up.
Some of the changes are:
- Single board design = smaller form factor
- Added Heatsink for on board cooling and spacing inside JD's head
- More even gain adjustment for lighting changes
- Flexible and modular ribbon connector for camera CCD + lens assembly
- Adjustable and/or removable lens, for focus or replacement lens
- RGB LED feedback
Version 2 of the camera is meant to match the operation of version 1 but with some quality of life improvements from a hardware and firmware perspective.
I used the standard JD example project. I added a script that made Servospeed 0 for the pan and tilt servos: D0 & D1 (The fastest speed). I also slowed down the camera frame rate a touch with the slider bar. The last thing that I changed was the vertical increment steps, I changed it to 2.
*Edit: I also adjusted the Grid lines to be closer to center
So just so I understand better:
*Set servo speeds to zero so servos will move as fast as possible.
*Slow down camera frame rate a little. (Why is this?)
*Changed was the vertical increment steps. (Why is this and will this setting be different for others?)
*Tighten up the grid lines. I assume this is because tracking takes place in the two side boxes that this gives the software more room to work trying to bring the object back to center?
Also, is the camera you're using in this demo stationary or mounted on the part of the robot that is panning back and forth?
Thanks again!
Also thanks for the tut on the tracking. I never thought about setting the speed first. That is fast and pretty smooth tracking. I look forward to trying it out with hard core Alan this week.
Correct, I wanted the servos to react as fast as possible.
I slowed down the camera frame rate a little bit to decrease the chance of missed frames due to communication slow downs over WiFi.
Yes, you are correct. In the tracking settings I changed the vertical increment steps to move a little farther with each movement, this will likely be different for each robots/servo. Distance away from the person being tracked will also affect this value.
Yes, you are correct again Dave :). I tightened up the grid lines so JD would have more time to detect my face moving away from center. The servos will only move if my face is detected in one of the 8 "boxes" (left, right, upper right, top, etc) surrounding the center box. To summarize, it gives JD more room and time to detect a change in face position.
The camera I was using was inside JD's head and his pan and tilt (neck and head) servos were doing the tracking.
I was using 3 horizontal incremental steps. I was connected via AP mode (no router involved)
I believe I was averaging 10fps.
Have you tried AP mode instead of client mode (connected to router)?
A direct connection (AP mode) is our default mode. It is very easy to connect via AP mode. Do you have anti-virus software that could be blocking your connection to ARC?
AP mode means your computer is connected directly to the ezb. You learned this when using the learn section for your ezrobot product.
Client mode means your ezb is connected to a router. You can only know of client mode if you followed the tutorial on the learn section for your ezrobot product.
There is no need for updating the connection tutorial. The last update of the tutorial was a detailed and fantastic video less than one week ago. There is no additional information. For client mode, only two peices of information is necessary: SSID and PASSWORD. Client mode is incredibly easy.
However, client mode depends on the speed of your router. Ezrobot does not own your router. We have no idea the speed of your router. The performance you experience in client mode has no relationship to ezrobot. The router speed is specific to the model and manufacturer, and not related to ezrobot.
If your computer is experiencing low frame rate with AP mode, then the pc speed may be the reason - OR the default settings have been changed.
1) Can you verify that you have the latest ARC?
2) why camera resolution setting are you using?
3) try adding a new camera control and enabling face tracking with ezrobot camera. Do NOT change any other settings. Do you still get a low frame rate?
First, every control has a ? (question mark). You can press that ? (question mark) to view help about that control. The control in question is the CAMERA CONTROL. Here is a direct link to the Camera Control manual page: http://www.ez-robot.com/Tutorials/Help.aspx?id=80
The reason for visiting that page is a section about VIDEO RESOLUTIONS. Please read that section. It is incredibly important that you understand what increasing the resolution does for a computer.
By increasing the video resolution to 640x480, look at the number of Bytes per second your CPU must process.
Lastly, i am curious about what other settings may have been modified in the camera control. Please visit my previous message and follow the instructions. Do not change any settings and use the camera control with Face tracking.
Everything I have typed is presented in the Camera Control manual page.