object-detection

问题 When i try run code into Jupyter notebook i getting Import error(attached image). I add paths to PYTHON_PATH and add %PYTHON_PATH% in system PATH, but i still get thos error 回答1: If you are using Anaconda , you must know that it ignores PYTHONPATH !. Use the following commands: conda develop ~/models/research/ conda develop ~/models/research/slim/ here is why you need to do it in this way. When you issue the above commands, it will create a .pth file inside your current's environment site

问题 I'm following tensorflow object detection api instructions and trying to train existing object-detection model("faster_rcnn_resnet101_coco") with my own dataset having 50 classes . So according to my own dataset, I created TFRecord (FOR training,evaluation and testing separately) labelmap.pbtxt Next, I edited model.config only for model-faster_rcnn-num_classes (90 -> 50(the number of classes of my own dataset), train_config-batch_size(1 -> 10), train_config-num_steps(200000 -> 100), train

问题 My objective is to have a sliding window slide over an image in overlapping steps so that I can run a classifier in each window and detect if an interesting object is there. For that, I need to make sure that windows I extract for classification truly do over the whole image, and grab the top and left coordinates of each sliding window on the original image. Following up from here: Sliding window - how to get window location on image? and based on this code for sliding windows: https://github

问题 My goal is to test a custom object-detection training using the Google ML-Engine based on the pet-training example from the Object Detection API. After some successful training cycles (maybe until the first checkpoint, since no checkpoint has been created) ... 15:46:56.784 global step 2257: loss = 0.7767 (1.70 sec/step) 15:46:56.821 global step 2258: loss = 1.3547 (1.13 sec/step) ... I received following error on several object detection training job trials: Error reported to Coordinator: , {

问题 Let say I want to train a HOG descriptor + Linear SVM for a car detection. Is it necessary for me to make, let say three classifiers, that are back-view, front-view and side-view of the car or I can just train a single classier for all viewpoints of the car? 回答1: It's not necessary but recommended. You can make a single classifier which handles multiple cases but it won't perform very well overall. The issue here isn't so much the variability of descriptor responses between the different

问题 I trained the cascade classifier to detect letters, this is the code I'm using. When I start the program, it opens my web cam but doesn't show the image (frames) and the detection but just a single blank window. I noticed this line causes the problem: faces_cascade.detectMultiScale(frame_gray, faces, 1.1, 2, 0 | CASCADE_SCALE_IMAGE, Size(30, 30)); This is the xml file. What should I do to make it show the captured frames and start detecting? Thank you. 回答1: I finally got it, I had to train my

问题 I'm working on a project at school, which basically is: writing an application to make a drone fly autonomously, and through scanning QR-codes hung up on walls, be able to navigate through a room in order to complete a certain task. What I am currently working on, is for the drone to detect cardboard boxes (working as obstacles). These boxes are white, and have a blue circle on them. How I'm planning to solve this, is by scanning the frame for colors and squares: If the drone detects a square

问题 Given the dataset of the object, I would like to extract that object from an image. The object is leaf in my case. It is easy in these kind of situation where there is only one big leaf in front of the camera. This can be done using the edge detected version of this picture as suggested in this answer as we are getting somewhat clear edge of what we want as output. for reference : But how can I extract all the leaves from an image in which there are a lot of such leaves. for example : for

问题 I use tensorflow object detection model to detect people on images,but there are　alse some televisions on images, and there are people on television. the model will detect people on television, how can i train the model to avoid to detect people on television, thanks. 回答1: The goal of most television is to present a lifelike image. It's probably impossible to detect if the pixel represents a real image or a TV image. What might work, depending a lot on your scenario, is to train a separate

问题 I'm trying to convert an object detection lst file to a rec file and train with it in SageMaker. My list looks something like this: 10 2 5 9.0000 1008.0000 1774.0000 1324.0000 1953.0000 3.0000 2697.0000 3340.0000 948.0000 1559.0000 0.0000 0.0000 0.0000 0.0000 0.0000 IMG_1091.JPG 58 2 5 11.0000 1735.0000 2065.0000 1047.0000 1300.0000 6.0000 2444.0000 2806.0000 1194.0000 1482.0000 1.0000 2975.0000 3417.0000 1739.0000 2139.0000 IMG_7000.JPG 60 2 5 12.0000 1243.0000 1861.0000 1222.0000 1710.0000