Aman Jaiswal

I am currently working as a Research Assistant in MANAS Lab, IIT Mandi in Prof Aditya Nigam's group, where I am working on gallbladder cancer (GBC) detection (in collaboration with Prof. Chetan Arora, IIT Delhi and Prof. Dr. Pankaj Gupta, SGPGI Chandigarh).

I also work as a Senior Machine Learning Engineer at Entrupy, where I work with the India-ML team to develop solutions for finerprinting and authenticating luxury/high-value goods. Prior to this, I worked as a Technological Account Manager at G7 CR Technologies.

I completed my Bachelor of Technology in Computer Science and Engineering (CSE) from Indian Institute of Technology Dharwad. I have had the opportunity to work with Prof Aditya Nigam in the summer of 2019. I have also collaborated with Mr. Vitobha Munigala of IBM Research Labs, Benagaluru under IBM's Global Remote Mentorship Programme.

I want to improve healthcare by making computers see, talk and act.

If you have any questions / want to collaborate / discuss new ideas, feel free to send me an email!

Google Scholar  /  Email  /  CV  /  Github  /  Twitter  /  LinkedIn

• [June 2024] Joined MANAS Lab, IIT Mandi as a Research Assistant (in Prof Adita Nigam's group).
• [June 2023] Patent published - Macroscopic fingerprint.
• [April 2023] Promoted to Senior Machine Learning Engineer at Entrupy.
• [August 2022] Patent granted - Monocular pose estimation and correction for sneaker authentication.
• [April 2022] Promoted to Machine Learnering Engineer II at Entrupy.
• [Sep 2020] Joined Entrupy as a Machine Learning Engineer.
• [May 2020] Joined G7CR Technologies as a Cloud Solution Architect.
• [April 2020] Graduated from IIT Dharwad.
• [Mar 2020] En-VStegNET accepted to IJCNN'20 (in association with IIT Mandi).
• [August 2019] Selected for IBM's Global Research Mentorship Programme.
• [May 2019] Started as a research intern at IIT Mandi.

We propose an Enhanced-VStegNET for full-video steganography, that outperforms the current SOTA VStegNET both quantitavily and qualitatively by modifying the architectures of the hiding and revealing networks that helps encode information more covertly and decode information more reliably.

Various embodiments of an apparatus, methods, systems and computer program products described herein are directed to a Fingerprint Engine that registers a reference image portraying a physical instance of an object. The Fingerprint Engine captures a query image portraying a physical instance of a target object. The Fingerprint Engine compares the reference image and the query image. The Fingerprint Engine determines an authenticity of the target object based on detecting a match between the reference image and the query image.

Various embodiments are directed to a Pose Correction Engine ( “ Engine ” ) . The Engine generates a reference image of the object of interest . The reference image portrays the object of interest oriented according to a first pose . The Engine receives a source image of an instance of the object . The source image portrays the instance of the object oriented according to a variation of the first pose . The Engine determines a difference between the first pose of the refer ence image and the variation of the first pose of the source image . The Engine identifies , based on the determined difference , one or portions of a three - dimensional ( 3D ) map of a shape of the object obscured by the variation of the first pose portrayed in the source image . The Engine generates a pose corrected image of the instance of the object that portrays at least a portion of the source image and at least the identified portion of the 3D map of the shape of the object.

Related Patent: Pose Estimation and Correction

• Cross-domain Matching: Worked on developing a recognition systems that can perform cross-sensor/cross-spectral IRIS identification.


• Unconstrained Ear Matching: Worked on developing a novel capsule-network based neural network architecture to perform ear-matching on the challenging UERC (Unconstrained Ear Recognition Challenge) dataset, improving upon the current state-of-the-art method for the same..


• Video Colorisation: Implemented and extended Pix2Pix, Colorful Image Colorization and other image colorization research papers to videos by using 3D CNN instead of 2D CNN and introducing a temporal loss component.

I borrowed this template from Jon Barron's website.