Hi @all, Is it possible to rotate camera lens by 90 degrees programmatically? We have below options, I don't see 90 DEG... class depthai.CameraImageOrientation Members: AUTO NORMAL HORIZONTAL_MIRROR VERTICAL_FLIP ROTATE_180_DEG Thanks & Best Regards, Ram

Hi @"ramkunchur"#p1578 , I do not think this is currently supported. The other rotations I believe are being done on by the image sensor itself (changing the readout/etc.). That said, I'm not sure if the Image Manip node can be used for this. Looking. Thanks, Brandon

On the subject of Image Manip - we will be adding quite a bit of functionality soon-ish, with the efforts below starting likely at the end of this month: https://github.com/luxonis/depthai/issues/376 That said, it looks like you can use Image Manip now to rotate an image 90 degrees: https://docs.luxonis.com/projects/api/en/latest/components/nodes/image_manip/ [URL=https://i.imgur.com/ihet5jd.png][IMG]https://i.imgur.com/ihet5jdh.png[/IMG][/URL] That said, I don't immediately see the function. Asking @"erik"#175 . Thanks, Brandon

Hello @"ramkunchur"#p1578 , here's a [minimal demo code](https://gist.github.com/Erol444/73383c07df936608bb7041790522dc16) that achieves rotating of the frame with ImageManip. You could also use `colorCamera` instead of mono camera - you would just need to change the values of the `dai.RotatedRect` (center and size) according to the resolution. I hope this helps! Thanks, Erik

Hi @"Brandon"#p1581 and @"erik"#p1584 . Thanks so much for the information... I'm trying to implement camera rotation to 90 degrees using specified method using colorCamera Below is code snippet, I get an error, seems there is unknown property error. pipeline = depthai.Pipeline() if camera: print("Creating Color Camera...") cam = pipeline.createColorCamera() # code to rotate camera by 90 degrees - start manip90 = pipeline.createImageManip() rr = depthai.RotatedRect() rr.center.x, rr.center.y = cam.getResolutionWidth() // 2, cam.getResolutionHeight() // 2 rr.size.width, rr.size.height = cam.getResolutionHeight(), cam.getResolutionWidth() rr.angle = 90 manip90.initialConfig.setCropRotatedRect(rr, False) cam.out.link(manip90.inputImage) # problem with this line... # code to rotate camera by 90 degrees - end cam.setPreviewSize(300,300) cam.setResolution(depthai.ColorCameraProperties.SensorResolution.THE_1080_P) cam.setInterleaved(False) cam.setBoardSocket(depthai.CameraBoardSocket.RGB) cam_xout = pipeline.createXLinkOut() cam_xout.setStreamName("cam_out") cam.preview.link(cam_xout.input) Error message as below: cam.out.link(manip90.inputImage) # problem with this line... AttributeError: 'depthai.ColorCamera' object has no attribute 'out' Requesting your kind help here.. Please let me know if my implementation is correct and also how to resolve this error. **I need to simply implement the following: The camera is placed horizontally, so need to rotate it to 90 degrees so that straight image is captured... This needs to happen both at sending image for inference and the one which we display finally in imshow method.** Thanks & Best Regards, Ram

Hello @"ramkunchur"#p1586, the [colorCamera](https://docs.luxonis.com/projects/api/en/latest/components/nodes/color_camera/) doesn't have `out` output, you should use `preview` instead. Please refer to our [documentation](https://docs.luxonis.com/projects/api/en/latest/components/nodes/color_camera/) for more information. I just tried it out with color camera instead of mono but it doesn't seem like it's working correctly ([code here](https://gist.github.com/Erol444/4659af8c53db5462673e3fb340622f9c)). We are investigating this further and it should be fixed this week, we apologize for the inconvenience. Thanks, Erik

Hi @"erik"#p1587 Thanks a lot for your quick response, really appreciate it. I'll wait for an update about the fix. Thanks & Best Regards, Ram

Hello @"ramkunchur"#p1588 , sorry for the troubles, it turns out I was incorrect - rotating color frames is already available, the width of the (input) frame just has to be in **multiples of 16**. Here's a [minimal example code](https://github.com/luxonis/depthai-python/blob/imagemanip-rotating-mono/examples/image_manip_rotate.py). Thanks, Erik

Hi @"erik"#p1591 , Thanks so much. I have come across another issue now, since I set preview to multiples of 16 i.e (640,400) originally it was (300,300) **cam.setPreviewSize(300,300)** changed to... **cam.setPreviewSize(640,400)** Inference doesn't work due to above changes, I tried to reset (300,300) to (640,400) in all other instances, inference doesn't work. Can you please tell me how I should adjust in other instances after changing preview size to 640,400? I also tried to change to (640,640) in all instances, get following error: **[NeuralNetwork(2)] [warning] Input image (640x640) does not match NN (300x300)** Need your help with this last step. Thanks & Best Regards, Ram

Hello @"ramkunchur"#p1593, here I quickly prepared a [minimal solution](https://gist.github.com/Erol444/4cead08da26adb1fddf4d7a8ec2d66e2) to achieve that. It uses another ImageManip to crop the frame into 300x300. You could also use cropping on the first imageManip, but for some reason, the output frame doesn't keep its aspect ratio (it gets stretched). Thanks, Erik

Hi @"erik"#p1596 Thanks so much for detailed help. The program doesn't seem to run... nothing happens Could you please check once? Thanks & Best Regards, Ram

Rotate camera by 90 DEGREES

erik

Hi chandrian ,
I assume you are using something similar to Lossless Zooming. So first you would want to rotate the frame 90deg (so people are upright), do the face detection, crop the original (rotated) 4k image into 1080x1920 (as in the lossless zooming example), then rotate that to 1080P, which you can feed into the UVC node. Thougths?
Thanks ,Erik

chandrian

Ok so this wouldnt be in the script then. I realize script is mostly for changing the pipeline anyway. Yes that sounds like a plan for me. I will attempt and let you know. Thanks!!

I will probably need to remove this before the rotate then? : cam.setVideoSize(1920, 1080)

erik

Hi chandrian , by default you will want to rotate the images by 90deg. So you will likely want 4k, then rotate it by 90deg, then do inference, then crop, then rotate back by -90deg to get to 1920x1080.

chandrian

Ok thanks! Is all of this happening in before the script node? Or is that unnecessary.

chandrian

And how does the script node work in terms of code path. I see a "while true" in the script with no breaks and a while true after the script. do they run in parallel?

chandrian

I tried keeping the same dimensions as my working code and just flipping twice and I didnt not get an output stream and then I tried a zero degree turn twice and still no stream. Am I messing something up here:

        manipRgb = pipeline.createImageManip()
        rgbRr = dai.RotatedRect()
        rgbRr.center.x, rgbRr.center.y = cam.getPreviewWidth() // 2, cam.getPreviewHeight() // 2
        rgbRr.size.width, rgbRr.size.height = cam.getPreviewHeight(), cam.getPreviewWidth()
        rgbRr.angle = 0
        manipRgb.initialConfig.setCropRotatedRect(rgbRr, False)
        cam.preview.link(manipRgb.inputImage)

        manipRgb2 = pipeline.createImageManip()
        manipRgb2.initialConfig.setCropRotatedRect(rgbRr, False)
        manipRgb.out.link(manipRgb2.inputImage)

        # Create an UVC (USB Video Class) output node. It needs 1920x1080, NV12 input
        uvc = pipeline.createUVC()
        manipRgb2.out.link(uvc.input)

chandrian

I actually cant get the cam.video to go through any manipulation node and into the UVC

chandrian

I tried passing the cam.video into the manip node and into the uvc. Then I tried setting the preview to 1920x1080 (is that a possible size?) and feeding that into the manip node and into uvc and I still could not get that working either.

erik

Hi chandrian ,
With the new depthai you can also use cam.video with ImageManip. I believe we plan to update the depthai uvc branch to latest, so you will be able to achieve this. Regarding the issue, please submit the full MRE.
Thanks, Erik

chandrian

Ok I will try to submit that. I have a deadline soon so I am not sure that will be done in time. Do you think it would be possible to rotate the facial recognition input so that, if the camera is 90 rotated, it will still recognize faces? ~~I will try that today but no luck so far.~~ Actually I think its working now.. more details to come
Thanks,
Aaron

edit:
~~Facial recognition seems to be working (blue square coming up) but not tracking at this moment.~~
edit2:
I think the blue squares were windows camera app tracking face, not the depthai.

chandrian

I am not having much success with rotating the input to the facial recognition. Do you think this is possible? If not, do you have another suggestion?

erik

Hi chandrian , yes it should be possible.

chandrian

Thanks for the reply Eric,

I have been trying to get this to work with no luck. It seems when I add multiple nodes, it does not function well. I am just feeding isp output into a mobilenet.

I tried just duplicating the resize image ImageManip above to prove functionality since I got the single resize working and I cannot get it to pass to the mobilnet and function.

This is using the UVC demo.

Thanks!

chandrian

Here's the code;

#!/usr/bin/env python3

import cv2
import depthai as dai
import blobconverter

# Create pipeline
pipeline = dai.Pipeline()

# Define source and output
camRgb = pipeline.create(dai.node.ColorCamera)
xoutVideo = pipeline.create(dai.node.XLinkOut)

xoutVideo.setStreamName("video")

# Properties
camRgb.setBoardSocket(dai.CameraBoardSocket.RGB)
camRgb.setResolution(dai.ColorCameraProperties.SensorResolution.THE_1080_P)
camRgb.setVideoSize(1920, 1080)

xoutVideo.input.setBlocking(False)
xoutVideo.input.setQueueSize(1)

# Create MobileNet detection network
mobilenet = pipeline.create(dai.node.MobileNetDetectionNetwork)
mobilenet.setBlobPath(
    blobconverter.from_zoo(name="face-detection-retail-0004", shaves=3)
)
mobilenet.setConfidenceThreshold(0.7)

# manipRgb = pipeline.createImageManip()
# rgbRr = dai.RotatedRect()
# rgbRr.center.x, rgbRr.center.y = camRgb.getPreviewWidth() // 2, camRgb.getPreviewHeight() // 2
# rgbRr.size.width, rgbRr.size.height = camRgb.getPreviewHeight(), camRgb.getPreviewWidth()
# rgbRr.angle = 0
# manipRgb.initialConfig.setCropRotatedRect(rgbRr, False)
#
#
# camRgb.isp.link(manipRgb.inputImage)
# manipRgb.out.link(mobilenet.input)



crop_manip2 = pipeline.create(dai.node.ImageManip)
crop_manip2.initialConfig.setResize(300, 300)
crop_manip2.initialConfig.setFrameType(dai.ImgFrame.Type.BGR888p)
camRgb.isp.link(crop_manip2.inputImage)
#crop_manip2.out.link(mobilenet.input)


crop_manip = pipeline.create(dai.node.ImageManip)
crop_manip.initialConfig.setResize(300, 300)
crop_manip.initialConfig.setFrameType(dai.ImgFrame.Type.BGR888p)
crop_manip.out.link(crop_manip2.inputImage)

# camRgb.isp.link(crop_manip.inputImage)
# crop_manip2.out.link(crop_manip.inputImage)
crop_manip.out.link(mobilenet.input)




# Script node
script = pipeline.create(dai.node.Script)
mobilenet.out.link(script.inputs["dets"])
script.outputs["cam_cfg"].link(camRgb.inputConfig)
script.outputs["cam_ctrl"].link(camRgb.inputControl)
script.setScript(
    """
    ORIGINAL_SIZE = (5312, 6000) # 48MP with size constraints described on IMX582 luxonis page
    SCENE_SIZE = (1920, 1080) # 1080P
    x_arr = []
    y_arr = []
    AVG_MAX_NUM=7
    limits = [0, 0] # xmin and ymin limits
    limits.append((ORIGINAL_SIZE[0] - SCENE_SIZE[0]) / ORIGINAL_SIZE[0]) # xmax limit
    limits.append((ORIGINAL_SIZE[1] - SCENE_SIZE[1]) / ORIGINAL_SIZE[1]) # ymax limit
    cfg = ImageManipConfig()
    ctrl = CameraControl()
    def average_filter(x, y):
        x_arr.append(x)
        y_arr.append(y)
        if AVG_MAX_NUM < len(x_arr): x_arr.pop(0)
        if AVG_MAX_NUM < len(y_arr): y_arr.pop(0)
        x_avg = 0
        y_avg = 0
        for i in range(len(x_arr)):
            x_avg += x_arr[i]
            y_avg += y_arr[i]
        x_avg = x_avg / len(x_arr)
        y_avg = y_avg / len(y_arr)
        if x_avg < limits[0]: x_avg = limits[0]
        if y_avg < limits[1]: y_avg = limits[1]
        if limits[2] < x_avg: x_avg = limits[2]
        if limits[3] < y_avg: y_avg = limits[3]
        return x_avg, y_avg
    while True:
    

        dets = node.io['dets'].get().detections
        if len(dets) == 0: continue
        coords = dets[0] # take first
        width = (coords.xmax - coords.xmin) * ORIGINAL_SIZE[0]
        height = (coords.ymax - coords.ymin) * ORIGINAL_SIZE[1]
        x_pixel = int(max(0, coords.xmin * ORIGINAL_SIZE[0]))
        y_pixel = int(max(0, coords.ymin * ORIGINAL_SIZE[1]))
        # ctrl.setAutoFocusRegion(x_pixel, y_pixel, int(width), int(height))
        # ctrl.setAutoExposureRegion(x_pixel, y_pixel, int(width), int(height))
        # Get detection center
        x = (coords.xmin + coords.xmax) / 2
        y = (coords.ymin + coords.ymax) / 2
        x -= SCENE_SIZE[0] / ORIGINAL_SIZE[0] / 2
        y -= SCENE_SIZE[1] / ORIGINAL_SIZE[1] / 2
        # node.warn(f"{x=} {y=}")
        x_avg, y_avg = average_filter(x,y)
        # node.warn(f"{x_avg=} {y_avg=}")
        cfg.setCropRect(x_avg, y_avg, 0, 0)
        node.io['cam_cfg'].send(cfg)
        node.io['cam_ctrl'].send(ctrl)
    """
)

# Linking
camRgb.video.link(xoutVideo.input)

# Connect to device and start pipeline
with dai.Device(pipeline) as device:

    video = device.getOutputQueue(name="video", maxSize=1, blocking=False)

    while True:
        videoIn = video.get()
        print("Done in seconds")


        # Get BGR frame from NV12 encoded video frame to show with opencv
        # Visualizing the frame on slower hosts might have overhead
        cv2.imshow("video", videoIn.getCvFrame())

        if cv2.waitKey(1) == ord('q'):
            break

erik

Hi chandrian ,
Please prepare the full MRE.
Thanks, Erik

chandrian

erik
Hi Erik,

What do you mean when you say that? Do I post those things here in the forum post? Or do I submit something like a git issue?

erik

Hi chandrian ,
You can submit it here, just make is minimal, as the code above isn't.
Thanks, Erik

chandrian

Ah ok thanks. I tried to keep it minimal with the images above but maybe that was too much? Basically I fed the camera.isp output through two image crops to see if it would work and I got an error. Both image crops worked independently but it feeding through both nodes gave me an error.

chandrian

what should the general approach be? This is the camera source flow through two nodes. The crop node works fine but I am not seeing the face recognition working when I add the rotation:

# Define source and output
camRgb = pipeline.create(dai.node.ColorCamera)
xoutVideo = pipeline.create(dai.node.XLinkOut)

xoutVideo.setStreamName("video")

# Properties
camRgb.setBoardSocket(dai.CameraBoardSocket.RGB)
camRgb.setResolution(dai.ColorCameraProperties.SensorResolution.THE_1080_P)
camRgb.setVideoSize(1920, 1080)
camRgb.setPreviewSize(300, 300)

xoutVideo.input.setBlocking(False)
xoutVideo.input.setQueueSize(1)

# Create MobileNet detection network
mobilenet = pipeline.create(dai.node.MobileNetDetectionNetwork)
mobilenet.setBlobPath(
    blobconverter.from_zoo(name="face-detection-retail-0004", shaves=3)
)
mobilenet.setConfidenceThreshold(0.7)


crop_manip = pipeline.createImageManip()
rgbRr = dai.RotatedRect()
rgbRr.center.x, rgbRr.center.y = camRgb.getPreviewWidth() // 2, camRgb.getPreviewHeight() // 2
rgbRr.size.width, rgbRr.size.height = camRgb.getPreviewHeight(), camRgb.getPreviewWidth()
rgbRr.angle = 90
crop_manip.initialConfig.setCropRotatedRect(rgbRr, False)
camRgb.isp.link(crop_manip.inputImage)

crop_manip2 = pipeline.create(dai.node.ImageManip)
crop_manip2.initialConfig.setResize(300, 300)
crop_manip2.initialConfig.setFrameType(dai.ImgFrame.Type.BGR888p)
crop_manip.out.link(crop_manip2.inputImage)

crop_manip2.out.link(mobilenet.input)

erik

Hi chandrian , I believe NNs should be run after rotating (90deg) the camera, as otherwise you will try to run inference on rotated frames (-90deg) which will have much worse performance?

« Previous Page Next Page »