Page 1

I am a second year computer science graduate student at IIIT-Hyderabad under the supervision of Dr.Madhava Krishna . My areas of interest are Machine learning, computer vision and robotics.

Earlier , I Completed my bachelors from BITS-PILANI with my thesis under Prof. Suman Kapur


• I will be joining MPI Tubingen as a intern under Dr Andreas Gieger

• One paper accepted to ICRA 2016, Stockholm, Sweden

• One paper accepted to ECMR 2015, Lincoln, UK

• One paper accepted to IROS 2015, Hamburg, Germany

• Two papers(oral) accepted to ICVGIP 2014(ACM),IISC ,banglore


Monocular Reconstruction of Vehicles: Combining SLAM with Shape Priors

Falak Chhaya, N Dinesh Reddy , Sarthak Upadhyay, Visesh Chari, M.Zeeshan Zia and K Madhava Krishna.  

IEEE International Conference on Robotics and Automation (ICRA),2016.

[Project Page]

Dynamic Body VSLAM with Semantic Constraints

N Dinesh Reddy , Prateek Singhal, Visesh Chari and K Madhava Krishna.  

IEEE International Conference on Intelligent Robots and Systems (IROS),2015.

[Project Page]

Semantic Motion Segmentation Using Dense CRF Formulation. 

N Dinesh Reddy, Prateek Singhal and K Madhava Krishna 

Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP), 2014 (ORAL) (~ 10% acceptance rate)

[Project Page] /[PDF] / [Slides] / [Videos] / [Bibtex]

Top Down Approach to Detect Multiple Planes from Pair of Images. 

Prateek Singhal, N Dinesh Reddy, Aditya Deshpande, Harit Pandya, K Madhava Krishna. 

Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP), 2014 (ORAL) (~ 10% acceptance rate)

[PDF] / [Slides] / [Videos] / [Bibtex]


selected projects

Driverless car challenge for Mahindra rise prize suitable for Indian roads

Under the supervision of Dr Madhava Krishna,IIIT Hyderabad and Dr.Shanti swarup medasani,URUMI Systems  

In this project, we are developing a completely autonomous car suitable for navigation in the indian road conditions. The perception for the autonomous car is achieved using low cost stereo camera sensors . The developments of the project are shared at this links

Facial expression detection on wild images using Active shape model

Under the supervision of Shailesh Kumar, Google INC  

Facial expressions convey non-verbal cues, which play an important role interpersonal relations. To increase the accuracy of facial expression detection , we have combined the active shape model with the gabor filter for better prediction. We have shown an improvement in the face detection by combining these features.

Improvement of Deformable parts model based on motion constraints

Under the supervision of CV Jawahar,IIIT Hyderabad  

Deformable parts model is a object detection algorithm , we have attempted to improve the accuracy of the object detection using motion constraints .We show that object detection is improved by adding motion constraints into the deformable parts.




My Google Scholar Page and YouTube Video's links:

GS             YT


109, EGK enclave

RTC 'x' Roads