Computer Vision & Machine Learning

Computer Vision & Machine Learning

Soliton provides automated image analysis solutions using the field of Computer Vision and its various sub-domains such as object detection, recognition etc. We use Machine Learning techniques to develop our Computer Vision solutions. Our Computer Vision work also includes using traditional techniques such as SVMs and Discrete Parts based Modelling along with the latest Deep Learning approaches.

Applications built on Computer Vision or Machine Learning systems must be extremely accurate. While prudent usage of Machine/Deep Learning systems help you attain high classification accuracies, giving a human level perception capability to the system raises the usability of the system by a few more notches. Our solutions achieve that by the meticulous fusion of contextual information with Deep Learning. We use Graph Algorithms and Probabilistic Graphical models.


  • Deep Neural Network solutions

    Solutions that address a variety of vision problems such as Object Detection / Recognition, Object Segmentation, Scene Understanding, Image Captioning etc.

  • Handwritten Character Recognition solutions

    Visualization of classified handwritten characters using deep neural networks. Our handwritten character recognition framework classifies characters with high accuracies. This, despite the challenges such as inherent variations in people’s handwritings and the unpredictability of string lengths.

Our Expertise

  1. Deep Learning for Visual Object Detection and Localization.
  2. Machine Learning approaches for non Deep Learning applications
  3. Segmentation Algorithms.
  4. Graph Algorithms and Probabilistic Graphical Models.
  5. Reinforcement Learning.
  6. Mathematical Optimization: Convex and Linear Programming.