object-detection

问题 I would like to train a deep learning framework (TensorFlow) for object detection with a new object category. As source for the ground truthing I have multiple video files which contain the object (only part of the image contains the object). How should I ground truth the video? Should I extract frame by frame and label every frame even when those video frames will be quite similar? Or what would be best practise for such a task? Open source tools are preferred. 回答1: It usually works as you

问题 In order to detect inselbergs from a geographic region, I've downloaded topographic imagery (relief and declivity) from that region. The declivity images seems best suited for the task. After applying a gaussian blur (or a much faster common blur, with ImageMagick), the image seems ready for automatic detection. Now I'm wondering the best/fastest way to detect these white stains on the black background. My first idea is to use a simple function (no external library) that works like the

问题 I have trained an SSD Mobilenet model with custom dataset(Battery). Sample image of the battery is given below and also attached the config file which I used to train the model. When the object is closer to the camera(tested with webcam) it detects the object accurately with probability over 0.95 but when I move the object to a longer distance it is not getting detected. Upon debugging, Found that the object gets detected but with the lower probability 0.35 . The minimum threshold is set to 0

问题 I have trained tensorflow object detection model (for num_steps:50000 ) using SSD (mobilenet-v1) on custom dataset. I got mAP@.50IOU ~0.98 and loss ~1.17. The dataset consist of uno playing card images (skip, reverse, and draw four). On all these cards, model performs pretty well as I have trained model only on these 3 card (around 278 images with 829 bounding boxes (25% bounding box used for testing i.e. validation) collected using mobile phone). However, I haven’t trained model on any other

问题 I'm looking to train my own object detector using YOLO 3 for a single class. Basically, it needs to detect whether the test images have the object or not. I face an error where the training doesn't begin and exits with an assertion '0' failed. I checked other answers which say that the CRLF must be encoded as LF for it to work on linux. But that solution doesn't work either. I'm following all the steps outlined in pjreddie's website! 回答1: The error happens due to image size or with batch size

问题 Hi I am trying to train SSD -mobilenet in-order to detect 13 classes. I also trained a faster rcnn -resnet101 . My training data images have resolution of 265 * 450 . (most of them) and each class had 400 images. Then this weird thing happened faster rcnn converged faster with batch size of 1 . But my SSD didn't . It's not converging at all . Here I will put the two graphs , Purple - Faster Rcnn | resnet101 Blue - SSD | mobilenet I have seen some have trained this for few classes and obtained

问题 For a performance measuring purpose I am trying to draw ROC curve. In ROC curve I have to plot False Positive Rate (FPR) in x-axis and True Positive Rate (TPR) in y-axis. As we know, FPR = FP/(FP+TN) So in the following picture how can i detect True Negative(TN) ? I have used HOG classifier to detect human. I marked with rectangle 1,2,3,4,5,6(or should be 7) to show the human objects that should be ignored and not to classify as human. and I think those are True Negative. In this picture i

问题 For a performance measuring purpose I am trying to draw ROC curve. In ROC curve I have to plot False Positive Rate (FPR) in x-axis and True Positive Rate (TPR) in y-axis. As we know, FPR = FP/(FP+TN) So in the following picture how can i detect True Negative(TN) ? I have used HOG classifier to detect human. I marked with rectangle 1,2,3,4,5,6(or should be 7) to show the human objects that should be ignored and not to classify as human. and I think those are True Negative. In this picture i

问题 I´m developing currently a webapp with object detection using the js-objectdetect libary. Therefore I´m converting opencv HAAR-cascades to the needed js-objectdetect format with a python script. My question is, if it´s possible to convert opencv LBP-cascades to the js-objectdetect format too. I need this, because I already have an opencv application in use and I need to use the same LBP-cascades. I don´t find anything on Javascript object-/ face detection with LBP-cascades on the internet.

问题 I am trying to create a haar cascade for hand detection and no matter what I do, I can't seem to get it finished. It always seems to end up stuck in some loop. I read a suggestion online that suggests to add code into the loop that will save the current working image into a temporary location. This solution would at least give me a place to start troubleshooting, but my problem here is my lack of C++ which OpenCV is written in. So if anyone can give me a suggestion of how to go about