Lets pimp the disco

first_imgTell us a bit about yourselves and how did you guys come together as a band?We’ve both been individual musicians for awhile now prior to this collaboration. Saba has been singing with a bunch of other bands as well as working as a professional sessions vocalist and playback singer, and Imaad is the frontman/ guitarist for The Pulp Society, as well as produces solo under the moniker Madboy. Imaad heard Saba’s voice at a late night jam with some friends and immediately knew he had to get her to sing on some of his productions. Also Read – ‘Playing Jojo was emotionally exhausting’And how did the name come about?It’s basically a combination of two solo projects. Madboy is Imaad’s project as a solo producer and Mink is Saba’s moniker as a producer and DJ. So we just thought combining the two would be simple and non-pretentious. Tell us about what drives you guys, what are your influences?Our personal influences are extremely varied and diverse but for this project we are focusing on a sound which is driven by old style Harlem swing and bringing that into context with production that’s rooted in disco and funk. It’s our own trademark sound and it’s hard to explain too much- it’s something that needs to be experienced. Also Read – Leslie doing new comedy special with NetflixTell us a bit about the songs you have planned for this chapter at Raasta?We haven’t been releasing too much of our material online, because we like each of our releases to be really special and when the EP is released, It’s going to be pretty hyped. But in spite of that, a lot of our tunes have been spreading by word of mouth and gigs that have travelled all over the place. Our songs – Pimp the Disco, Taste your kiss, Lemonade (which was our first single release and is on Soundcloud) Trouble, Soulboy etc and many more will all be on the set list. What’s next in the pipeline for you guys?We play Sunburn in Goa which is the last gig for the year for us coming at the end of our first year as ‘Madboy/Mink’ and it’s been a very busy year in terms of gigs. We’re just going to take stock, lock ourselves in the studio for a little while, and complete the final mixes for our EP which will be out most probably by the end of Feb or beginning of March – and will be followed by a comprehensive tour across the country.WHEN: 25 December, 9 pm onwardsWHERE: Raasta, Hauz Khas Villagelast_img read more

Facebook introduces Rosetta a scalable OCR system that understands text on images

first_imgYesterday, researchers at Facebook introduced a machine learning system named, Rosetta for scalable optical character recognition (OCR). This model extracts text from more than a billion public Facebook and Instagram images and video frames. Then, this extracted text is fed into a text recognition model that has been trained on classifiers, which helps it understand the context of the text and the image together. Why Rosetta is introduced? Rosetta will help in the following scenarios: Provide a better user experience by giving users more relevant photo search results. Make Facebook more accessible for the visually impaired by incorporating the texts into screen readers. Help Facebook proactively identify inappropriate or harmful content. Help to improve the accuracy of classification of photos in News Feed to surface more personalized content. How it works? Rosetta consists of the following text extraction model: Source: Facebook Text extraction on an image is done in the following two steps: Text detection In this step, rectangular regions that potentially contain the text are detected. It performs text detection based on Faster R-CNN, a state-of-the-art object detection network. It uses Faster R-CNN but replaces ResNet convolutional body with a ShuffleNet-based architecture for efficiency reasons. The anchors in regional proposal network (RPN) are also modified to generate wider proposals, as text words are typically wider than the objects for which the RPN was designed. The whole detection system is trained jointly in a supervised, end-to-end manner. The model is bootstrapped with an in-house synthetic data set and then fine-tuned with human-annotated data sets so that it learns real-world characteristics. It is trained using the recently open-sourced Detectron framework powered by Caffe2. Text recognition The following image shows the architecture of the text recognition model: Source: Facebook In the second step, for each of the detected regions a convolutional neural network (CNN) is used to recognize and transcribe the word in the region. This model uses CNN based on the ResNet18 architecture, as this architecture is more accurate and computationally efficient. For training the model, finding what the text in an image says is considered as a sequence prediction problem. They input images containing the text to be recognized and the output generated is the sequence of characters in the word image. Treating the model as one of sequence prediction allows the system to recognize words of arbitrary length and to recognize the words that weren’t seen during training. This two-step model provides several benefits, including decoupling the training process of detection and recognition models, recognition of words in parallel, and independently supporting text recognition for different languages. Rosetta has been widely adopted by various products and teams within Facebook and Instagram. It offers a cloud API for text extraction from images and processes a large volume of images uploaded to Facebook every day. In future, the team is planning to extend this system to extract text from videos more efficiently and also support a wide number of languages used on Facebook. To get a more in-depth idea of how Rosetta works, check out the researchers’ post at Facebook code blog and also read this paper: Rosetta: Large Scale System for Text Detection and Recognition in Images. Read Next Why learn machine learning as a non-techie? Is the machine learning process similar to how humans learn? Facebook launches a 6-part Machine Learning video serieslast_img read more