Welcome!
Hi!
I'm Abbhinav Venkat, an ML Research at StarMall Inc, an MIT founder's early stage startup that works in the furniture e-commerce space.
Previously, I was a Research Assistant at the Geometric Computing Lab at Stanford University, working with Prof. Leonidas Guibas and Prof. Srinath Sridhar on modeling real-world dynamics from videos.
I've completed by BTech and MS (by Research) at the Centre for Visual Information Technology (CVIT), International Institute of Information Technology, Hyderabad (IIIT-H), under the guidance of Dr. Avinash Sharma. My expertise lies at the intersection of Machine Learning and 3D.
Scroll below, for latest updates! To know more about my past research projects, click here!
I'm a fitness enthusiast and enjoy writing poetry as well. Visit my personal blog to know my story!
Recent News
Nov' 22. Joined StarMall, an early stage Silicon Valley startup that works in the furniture e-commerce space.
Aug' 20. Joined Uplara, an MIT founder's early stage startup that works on Virtual Clothing Tryon, as a Computer Vision Scientist.
July' 20. Completed MS Thesis on Monocular Human Body 3D Reconstruction
March '20. Completed Research stint at Stanford
Sep '19. Paper on "DeepHuMS: Deep Human Motion Signatures for Skeletal Sequences" accepted at ACPR, 2019 (Oral; < 10% acceptance rate).
Aug '19. Paper on "HumanMeshNet: Polygonal Mesh Recovery of Humans" accepted at ICCV-W, 2019.
June '19. Research Assistant at Stanford University with Prof. Leonidas Guibas.
Dec '18. BMVC 2018 paper presented at "Vision India" Track at ICVGIP 2018, Hyderabad.
Nov '18. Gave a talk at the Max Planck Institute for Informatics (MPI-Inf, Saarbrücken) on my BMVC paper.
Oct '18. Received Silver Medal in college's 5k race.
July '18. Paper on "Deep Textured 3D Reconstruction of Human Bodies" accepted at BMVC, 2018.
May '18. Won Qualcomm's Innovation Fellowship (QInF) for our proposal on "Textured Non-rigid 3D Reconstruction in a calibration free environment"
2014 - 2017. Dean's List at IIIT-Hyderabad, for being in the top 3% of my undergraduate batch.
Jan '18. Received Silver medal in college's 10k race.
Research Interests
We live in a 3-Dimensional world. Our eyes are able to interpret the volume of objects, relative positioning between them, their categories, possible deformations, texture and much more, from the light rays that enter them. My research attempts at mimicking the human visual system, for the specific case of interpreting the 3D of non-rigid objects (such as humans, birds etc.) from RGBD images. Further, I'd like to perform higher order analysis, given the 3D of the object.
Going ahead, I'd like to specialise in Robotics, Applied ML and Computer Vision. In addition to my existing work, my research interests extend to broader topics such as:
Behaviour understanding from 3D (for e.g: analysing the body language of humans)
Analysing human-object and human-human interactions
3D Scene Understanding