There are 2 things i want to clarify
Why is it that when i open tensorboard for eval file after training using object detection api, i see 2 of the same test image side by side? Does it mean that each picture goes through detection twice? I have attached an example below.
Test image
I only can view 9 test images in tensorboard but I have 16 test images. Is there a way i can view all 16 test images?
You see two images side by side. The right image is your labeled image as groundtruth. The left image is your prediction result. Why only 9 test images is visible I can't tell you for now.
Related
UseCase: I'm trying to extract certain parts of a screenshot which is taken from a game (with a tf object detection model) and extract the text within this part (custom model for the font used in the game).
I have trained a custom model based on SSD Mobilenet V2 and the object detection works quite okish, but sometimes the bounding box is off. I googled about selecting the right images and the right amount for training the custom model, but I couldn't find a good hint in the right direction.
I try to extract the following (surrounded by red):
The environmen can change:
Resolution of the game can be different (1920x1080, WHQD etc.)
Text in the box is not always the same
I have trained with 120 self made images (1920x1080) (90% for training 10% for test) (all of these images where a screenshot of the game) and as I mentioned the results are okish. Sometimes the detected area is off (cutting the content of the box or including a lot area of the box surroundings).
Maybe someone can help me/answering the following questions:
Could a bigger training dataset increase the accuracy?
Should I also take different resolutions into account when creating the training data?
Would it make sense to feed only the boxes without the rest of the game screenshot into the training? Or should I mix screenshots of the whole game and only box screenshots?
Thank you in advance ! :)
The ILSVRC 2012 small classification dataset is not separated by folder and don't have a labels file. How get the labels for the training set?
I tried on nonpub downloads page but does not exist anymore, and i tried by the filenames but their don't have the synset id on it.
I've been having the same issue today following this tutorial on reproducing ImageNet Validation results. I think I've found an answer, even if partial
In the article they point out to this link to get the validation set for object detection. I downloaded it and had the same issue as yourself, it only contains images without labels. What I've found is that this same website had this other link for the bounding boxes. I've downloaded it and alongside with the bboxes it comes with the proper class for each image
Hope this helps!
I have 1600 videos and I want to make joint annotation label data about videos.
I've already made the open pose network and I put my videos as input of the network and saved the joint data as json file.
When I put my first video data as input, there are no errors. And when I put second, third video as input, there are no errors too.
But When I put the fourth video data as input, I got these error message.
enter image description here
enter image description here
these above images are the error message.(OOM)
The size of first, second, third, fourth video is the same.
When I change name first and fourth video name, I got the same error when putting fourth video.
I think this error is about the graph. but I couldn't know why exactly.
I think there are many genious on stackoverflow. So please answer my question... :)
I solve this problem by using cpu. not using gpu.
I use cpu only in tensorflow for solving this problem. and it works!
Is it possible to view the images for all steps in the Image viewer of tensorboard?
Only the images corresponding to the last step are shown in the Images tab, though they are accessible changing the index of the the url of the image:
(Changing the 0 for the required step in the url: localhost:6006/data/individualImage?index=0&tag=777_mask_output_output_1%2Fimage%2F0&run=tensorboard_logs&ts=1487721119.455751)
As I understood, an additional slider was added at TensorFlow v. 1.1.0:
I have 1 image which consists of 4 different images which I need to copy and save it as four different images. Is there any easy way to do it as I have more than 100 such images.
Can you supply a dropbox link to one of the images so I can see what you mean? Do you mean you have one scan of 4 images all on the same layer, and you need to crop 4x and resave individual images? Or is each image on its own layer?
Are the 100 images all exactly the same layout? If so, you may be able to create an action that would cut down on your time.
If the images are on separate layers, you can use File > Scripts > Export Layers to files...