Can I use detect.py with --batch 4 (4 USB webcams)?

[emasson-maxx]

Hi there,
I was wondering if I could easily scale my Yolov5 detection from a single webcam to 4 with the --batch 4 --source <file containing a list of /dev/videoX> ?

Just starting with Yolov5 and I love it.
Eric

[glenn-jocher]

@emasson-maxx yes you put your stream addresses in a *.txt file with one per line, no commas, and pass the *.txt file as --source *.txt

[glenn-jocher]

batched inference will be run automatically with batch size equal to the number of streams

[emasson-maxx]

So I could define my streams.txt as followed?
/dev/video1
/dev/video3
/dev/video5
/dev/video7

[emasson-maxx]

Also, there will be a single Yolo model loaded into memory, right?

[glenn-jocher]

Yes

[SwHaraday]

Hi !
If you want detect 4 images at once, try below.

class Load4Streams:
    # Can be used instead of LoadStreams in Datasets.py
    def __init__(self, sources='streams.txt', img_size=640, stride=32, auto=True):
        self.mode = 'stream'
        self.img_size = img_size
        self.stride = stride

        if os.path.isfile(sources):
            with open(sources) as f:
                sources = [x.strip() for x in f.read().strip().splitlines() if len(x.strip())]
        else:
            sources = [sources]

        print(sources)
        n = len(sources)
        self.imgs, self.fps, self.frames, self.threads = [None] * n, [0] * n, [0] * n, [None] * n
        self.sources = [clean_str(x) for x in sources]  # clean source names for later
        self.auto = auto
        for i, s in enumerate(sources):  # index, source
            # Start thread to read frames from video stream
            st = f'{i + 1}/{n}: {s}... '
            if 'youtube.com/' in s or 'youtu.be/' in s:  # if source is YouTube video
                check_requirements(('pafy', 'youtube_dl'))
                import pafy
                s = pafy.new(s).getbest(preftype="mp4").url  # YouTube URL
            s = eval(s) if s.isnumeric() else s  # i.e. s = '0' local webcam
            cap = cv2.VideoCapture(s)
            #assert cap.isOpened(), f'{st}Failed to open {s}'
            w = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
            h = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
            self.fps[i] = max(cap.get(cv2.CAP_PROP_FPS) % 100, 0) or 30.0  # 30 FPS fallback
            self.frames[i] = max(int(cap.get(cv2.CAP_PROP_FRAME_COUNT)), 0) or float('inf')  # infinite stream fallback

            if cap.isOpened():
                _, self.imgs[i] = cap.read()  # guarantee first frame
                self.threads[i] = Thread(target=self.update, args=([i, cap, s]), daemon=True)
                LOGGER.info(f"{st} Success ({self.frames[i]} frames {w}x{h} at {self.fps[i]:.2f} FPS)")
                self.threads[i].start()
                #print('** ', self.threads) # debug print
            else:
                print(f'{st}Failed to open Cam {s}')
                self.imgs[i] = np.full((480, 640, 3), (128, 128, 128), dtype=np.uint8)

        LOGGER.info('')  # newline
        # ここで4つの画像を合成する
        self.concimg = cv2.hconcat([self.imgs[0], self.imgs[1]])
        if len(sources) > 2:
            conc2 = cv2.hconcat([self.imgs[2], self.imgs[3]])
            self.concimg = cv2.vconcat([self.concimg, conc2])
            self.concimg = cv2.resize(self.concimg, (800, 600), interpolation = cv2.INTER_AREA)
        else:
            self.concimg = cv2.resize(self.concimg, (800, 300), interpolation = cv2.INTER_AREA)
        self.obi = np.full((20, 800, 3), (255, 255, 255), dtype=np.uint8)
        self.concimg = cv2.vconcat([self.concimg, self.obi])
        self.concimg = np.expand_dims(self.concimg, axis=0) # CHW > BCHW 

        # check for common shapes
        #s = np.stack([letterbox(x, self.img_size, stride=self.stride, auto=self.auto)[0].shape for x in self.imgs])
        s = np.stack([letterbox(x, self.img_size, stride=self.stride, auto=self.auto)[0].shape for x in self.concimg])
        self.rect = np.unique(s, axis=0).shape[0] == 1  # rect inference if all shapes equal
        if not self.rect:
            LOGGER.warning('WARNING: Stream shapes differ. For optimal performance supply similarly-shaped streams.')

    def update(self, i, cap, stream):
        # Read stream `i` frames in daemon thread
        n, f, read = 0, self.frames[i], 1  # frame number, frame array, inference every 'read' frame
        while cap.isOpened() and n < f:
            n += 1
            # _, self.imgs[index] = cap.read()
            cap.grab()
            if n % read == 0:
                success, im = cap.retrieve()
                if success:
                    self.imgs[i] = im
                else:
                    LOGGER.warning('WARNING: Video stream unresponsive, please check your IP camera connection.')
                    self.imgs[i] = np.zeros_like(self.imgs[i])
                    cap.open(stream)  # re-open stream if signal was lost
            time.sleep(1 / self.fps[i])  # wait time

    def __iter__(self):
        self.count = -1
        return self

    def __next__(self):
        self.count += 1
        #if not all(x.isAlive() for x in self.threads) or cv2.waitKey(1) == 27: #ord('q'):  # q to quit
        if cv2.waitKey(1) == 27:#ord('q'):  # q to quit 
            cv2.destroyAllWindows()
            raise StopIteration
        # ここで4つの画像を合成する
        self.concimg = cv2.hconcat([self.imgs[0], self.imgs[1]])
        if len(self.sources) > 2:
            conc2 = cv2.hconcat([self.imgs[2], self.imgs[3]])
            self.concimg = cv2.vconcat([self.concimg, conc2])
            self.concimg = cv2.resize(self.concimg, (800, 600), interpolation = cv2.INTER_AREA)
        else:
            self.concimg = cv2.resize(self.concimg, (800, 300), interpolation = cv2.INTER_AREA)
        self.obi = np.full((20, 800, 3), (255, 255, 255), dtype=np.uint8)
        self.concimg = cv2.vconcat([self.concimg, self.obi])
        self.concimg = np.expand_dims(self.concimg, axis=0) # CHW > BCHW 

        # Letterbox
        img0 = self.concimg.copy()
        img = [letterbox(x, self.img_size, stride=self.stride, auto=self.rect and self.auto)[0] for x in img0]

        # Stack
        img = np.stack(img, 0)

        # Convert
        img = img[..., ::-1].transpose((0, 3, 1, 2))  # BGR to RGB, BHWC to BCHW 
        img = np.ascontiguousarray(img)

        return self.sources, img, img0, None, ''

    def __len__(self):
        return len(self.sources)  # 1E12 frames = 32 streams at 30 FPS for 30 years