Hi, for your project, I will apply tweepy extracted tweets on python text-to-image, Image and ImageDraw libraries or use existing github code on the tweets extracted by tweepy and then convert the tweets into images.
Also I can also apply interesting feature like in addition to simple text, I will apply tweet image with text and convert these total into a single image or separate image if you like as a value addition to your project.
I can do advanced, NLP and computer vision feature like specific parts of tweets marking if you want.
I will also add a lot of lucid, accurate and in-detail documentation in the code so that it can understood easily.
As a mention of my deep learning computer vision projects experience-
1)
I have coded a project on Twitter tweets live extraction in REAL TIME and convert each tweets to images using tweepy library of python.
2)
I have coded a project on Twitter tweets live extraction in REAL TIME and done the spell checking and done ratings on different tweets based on emotion type using textblob, texttable and NLTK libraries of python in Deep learning NLP or natural language processing code and then graphically marked the emotion words by converting the selected tweets into image in python.
3)
I have done a handwriting recognition project in an image using Tesseract library of python.
4)
I have already done an image extraction and classification project using deep learning neural network in tensorflow and keras and got more than 96 % accuracy in this project by evaluating the accuracy and loss on training data and also on validation data for analysing data overfitting and underfitting.
5)
I have coded a GAN(more specifically DCGAN) code for human face generation using deep learning neural network code in python in my google colaboratory python notebook.
6)
I have completed a body pose estimation project in python using deep learning neural network and added some custom graphics in it by changing some if the default graphics.
7)
I have coded a computer vision project by changing default YOLO 3 sbd OpenCV code with custom requirements.
In this project, in a soccer video to identify person and object, I have
--> added 2 types of custom programmed graphics and
--> even changed the location graphics mapping,
--> removed default YOLO 3 rectangles and labels and
--> even changed labels in the default YOLO 3 and OpenCV code.
8)
I have done a project on human face emotion identification using tflearn and other python libraries.
Feel easy to contact me further for your question and comment and I will be happy to help you in solving your issue or task successfully.
Thank you for your time.
yours sincerely,
Pritam