I want to classify food and non-food image by using Clarifai API.
It seems that the API is trying to assume that all images are food images, by default. For example, U pushed an image of people or animals to Clarifai and return back to us the results "water, beer..", etc with very high probability. Is there any way to overcome this problem?.
‘Food’ model is able to analyze images and video around food (even down to the ingredient level)! It’s more speciifc and narrow, while ‘General’ model is used to analyze a more wide range of visual content.
Related
I want to build an image generator model for interior room design.
This model should be able to generate an interior image of living room, bedroom, hall, kitchen, bathroom etc.
I have searched about it and found out the following websites.
https://interiorai.com/
https://image.computer/
And I made this picture when visiting https://image.computer.
[! Contemporary living room with a large TV and 3 windows]1
Above result was perfectly what I want. But free account was restricted to 10 credit images.
And input data don't have to be sentenced, just options are enough for me (e.g. Style: modern, type: living, equipments: [TV: wide, Curtain: white, Window: 3]).
So I decided to google pretrained model of interior design generator, and finally gave up.
I would like to build a tensorflow(or keras) model that acts just like image.computer. Please let me find a model or build model.
Thanks
So, I run a website which has a card database for the old Star Wars Trading Card Game by Wizards of the Coast. swtcg.com There are multiple sets/expansions and each of those has multiple cards.
If you google other trading card games like Magic the Gathering or Pokemon TCG, you SOMETIMES will get rich, carousel-style results for individual cards, and if you click one of the cards, you get the rich, graph sidebar result. It seems like google is aware that these are Cards from Sets for a Trading Card Game.
I have tried to search for sites that are using structured data to identify these types, but have only found one or two, and they are just using Product markup.
Does anyone have any advice for what types I should use? I would really like to get to the point where you could search for a card and could get a rich result on the side with details about each card.
I've tried Product, but only some of them are cards that are actually sold. Others are digital only and free. I've considered Article and Creative Work, but am just really stumped as to what the best options would be for me. Is there such a thing as custom types that aren't insanely difficult to implement?
The contents of your website present a database for playing cards. Let's look at your web page representing one card 100 Battle Droids. In my humble opinion, this content is explicit Creative Work and this type can probably help you. Due to the fact that the subject of this web page is a game, the use of the embedded type Game can help you. For this type, you can use the about or mainEntity properties as alternatives.
The map that is presented in the content of this web page is an image. You can probably be helped by using the following Google guidelines for structured data for the Article type:
For best results, provide multiple high-resolution images (minimum of
300,000 pixels when multiplying width and height) with the following
aspect ratios: 16x9, 4x3, and 1x1. For example:
{
"#context": "https://schema.org",
"#type": "NewsArticle",
"image": [
"https://example.com/photos/1x1/photo.jpg",
"https://example.com/photos/4x3/photo.jpg",
"https://example.com/photos/16x9/photo.jpg"
]
}
You can use the free online calculator for the following aspect ratios:
4X3 and 16x9. To compress your images, you can search for image compressors on the web. I usually use Compressnow with the maximum level of 65%.
Using the Google Guides to optimize your images Google Image best practices and UX to responsive images.
Your information below the card is the table. The use of a responsive table (row only) for this data may probably help. You can use the W3 guide Generating JSON from Tabular Data on the Web to structure this data.
You can use Google guide to Dataset and the standard of W3 Data Catalog Vocabulary (DCAT) - Version 2 to create a database of your cards.
I am new to iOS swift programming and to building AR apps with ARKIT. I find that ARKIT is more powerful than I imagined and I can able to achieve all my business case but except placing data dashboards or charts in AR 3d space. I found ARCharts on Google but it seems to be useless.
My business case is actually scan the object or product and recognize it and display data related to it on AR world which should also show some data analytics dashboard for sales trends of the product.
How to achieve this.. pls provide some inputs
Using ARKit you will be able to detect image detection, object detection and plane detection. For your business case, you can use image detection and object detection.
I will recommend you to go through the below tutorial to get some basic knowledge on object detection and image detection.
Building a Museum App with ARKit 2. Happy coding ;)
I'm relatively new and am still learning the basics. I've used NVIDIA DIGITS in the past, and am now looking at Tensorflow. While I've been able to fumble my way around creating some models for a few projects I'm working on, I really want to start diving deeper into what I'm doing, how I'm doing it, and ultimately a better understanding of why.
One area that I would like to start with is the Images that I'm using for training and testing. Can anyone point me to a blog, an article, a paper, or give me some insight in what I need to consider when selecting images to train a new model on. Up until recently, I've been using datasets that have already been selected and that are available for download. Lets say I'm going to start working on a project that involves object detection of ships from a variety of distances and angles.
So my thoughts would be
1) I need a large quantity of images.
2) The images need to contain ships of the different types I would like to detect. (lets just say one class, ships, don't care what type of ships)
3) I also need to have images that have a great variety of distance perspective for the different types of ships.
Ultimately, my thoughts are that the images need to reflect the distance, perspective, and types of ships I would ideally want to identify from the video. Seems simple enough.
However, there are a number of questions
Does the images need to be the same/similar resolution as the camera I'll be using, for best results?
Does the images all need to be the same resolution?
Can I use a single image and just digitally zoom out on the image to give the illusion of different distances?
I'm sure there are a number of other questions that I'm not asking, or should be asking. Are there any guide lines available for creating a solid collection of images to use when creating the collection of images for training and validation?
I recommend thinking through end to end, like would you need to classify ship models as a next step? I recommend going through well known public datasets and actually work with the structure, how to store data, labels, how to handle preprocessing etc.
More importantly, what are you trying to achieve? Talking to experts in the topic does help greatly while preparing your own dataset.
Use open source images if you can, e.g. flickr, google, imagenet.
No, you don't need them to be the same resolution.
It is not ideal to zoom in/out images to use in different categories. Preprocessing images and data augmentation already does this to create more distant representations of the same class. This is why I would recommend hands on approach with an existing dataset first.
Yes, what you need is many, different representations of classes, and a roughly balanced dataset of classes. If you define your data structure well in the beginning, it will save you a ton of time as you won't have to make changes often.
I am training a classifier for recognizing certain objects in an image. I am using the Watson Visual Recognition API but I would assume that the same question applies to other recognition APIs as well.
I've collected 400 pictures of something - e.g. dogs.
Before I train Watson, I can delete pictures that may throw things off. Should I delete pictures of:
Multiple dogs
A dog with another animal
A dog with a person
A partially obscured dog
A dog wearing glasses
Also, would dogs on a white background make for better training samples?
Watson also takes negative examples. Would cats and other small animals be good negative examples? What else?
You are right that this is a general issue for all kinds of custom classifiers and recognizers - be it vize.it, clarifai, IBM Watson, or training a neural network on your own say in caffe. (Sorted by the number of example images you need to use.)
The important thing you need to ask is how are you going to use the classifier? What are the real images you will feed the machine to predict the objects shown? As a general rule, your training images should be as similar to predict-time images as possible - both in what they depict (kinds and variety of objects) and how they depict it (e.g. backgrounds). Neural networks are super-powerful and if you feed them enough images, they will learn even the hard cases.
Maybe you want to find dog images in user's folders - which will include family photos, screenshots and document scans. Reflect that variety in the training set. Ask the user if a dog with another animal should be tagged as a dog photo.
Maybe you want to find dog images on a wilderness photo trap. Just use various images taken by that photo trap (or several photo traps, if it's a whole network).
In short — tailor your sample images to the task at hand!