Maneesh Bilalpur

I'm a PhD candidate at the University of Pittsburgh's School of Computing and Information, where I work on Multimodal Machine Learning. I am supervised by Prof. Jeff Cohn. My interests include utilizing facial expressions, voice prosody, body pose, and speech for applications in affective computing. I am also curious about how these multimodal behaviors shape day-to-day human communication.

Prior to my PhD, I've worked at a computer vision data annotation startup Playment on semantic segmentation and object detection. My Masters of Science thesis was supervised by Prof. Ramanathan Subramanian, and Prof. Mohan Kankanhalli at iiit hyderabad.

Email / CV / Scholar / Twitter / LinkedIn / Github

News

Presented my dissertation work at the Doctoral Consortium at FG 2025.
Awarded the Provost's Dissertation Completion Fellowship for Fall 2025 at Pitt.
Delivered a guest lecture on Multimodal LLMs for the CS 2750 course at Pitt.
Received the Oustanding Reviewer Award at FG 2024.

Research

My experience of multimodal machine learning research spans multiple domains including Computer Vision, Speech Processing, and Natural Language Processing. I also tinker with statistics (mixed-effects) for hypothesis testing in human behavior. Here are some papers that I have worked on. The exhaustive list can be found on my google scholar profile.

	Language use in depressed and non-depressed mothers and their adolescent offspring Laura A Cariola, Lisa B Sheeber, Nicholas Allen, Maneesh Bilalpur, Timothy Bird, Saurabh Hinduja, Louis-Philippe Morency, and Jeffrey F Cohn Journal of Affective Disorders (Impact Factor: 4.9), 2024
	Expanding PyAFAR: A Novel Privacy-Preserving Infant AU Detector Itir Onal Ertugrul, Saurabh Hinduja, Maneesh Bilalpur, Daniel S Messinger, and Jeffrey F Cohn IEEE FG, 2024
	Learning to generate context-sensitive backchannel smiles for embodied ai agents with applications in mental health dialogues Maneesh Bilalpur, Mert Inan, Dorsa Zeinali, Jeffrey F Cohn, and Malihe Alikhani AAAI Workshop, 2024
	Argumentative stance prediction: An exploratory study on multimodality and few-shot learning Arushi Sharma, Abhibha Gupta, and Maneesh Bilalpur* (equal contributions) EMNLP Workshop on Argument Mining*, 2023
	SHAP-based Prediction of Mother's History of Depression to Understand the Influence on Child Behavior Maneesh Bilalpur, Saurabh Hinduja, Laura Cariola, Lisa Sheeber, Nicholas Allen, Louis-Philippe Morency, and Jeffrey F Cohn ACM ICMI, 2023
	PyAFAR: Python-based automated facial action recognition library for use in infants and adults Saurabh Hinduja, Itir Onal Ertugrul, Maneesh Bilalpur, Daniel S Messinger, and Jeffrey F Cohn IEEE ACII Demo, 2023
	Infant AFAR: Automated facial action recognition in infants Itir Onal Ertugrul, Yeojin Amy Ahn, Maneesh Bilalpur, Daniel S Messinger, Matthew L Speltz, and Jeffrey F Cohn Behavior Research Methods, 2023
	Multimodal Feature Selection for Detecting Mothers' Depression in Dyadic Interactions with their Adolescent Offspring Maneesh Bilalpur, Saurabh Hinduja, Laura A Cariola, Lisa B Sheeber, Nick Alien, László A Jeni, Louis-Philippe Morency, and Jeffrey F Cohn IEEE FG, 2023
	Ballistic Timing of Smiles is Robust to Context, Gender, Ethnicity, and National Differences Maneesh Bilalpur, Saurabh Hinduja, Kenneth Goodrich, and Jeffrey F Cohn IEEE ACII, 2022
	Gender and emotion recognition with implicit user signals Maneesh Bilalpur, Seyed Mostafa Kia, Manisha Chawla, Tat-Seng Chua, and Ramanathan Subramanian ACM ICMI, 2017
	Discovering gender differences in facial emotion recognition via implicit behavioral cues Maneesh Bilalpur, Seyed Mostafa Kia, Tat-Seng Chua, and Ramanathan Subramanian IEEE ACII, 2017

Work Experience

Playment (acquired by Telus International)

Computer Vision Researcher
Bengaluru, India · Aug 2018 – Jul 2019

Developed and deployed interactive semantic segmentation models (including raster-to-vector conversion) to accelerate annotation. Standardized 3D point clouds and implemented 3D→2D projections for a widely used annotation tool in self-driving applications. Shipped a GDPR-compliant anonymizer that automatically blurs faces and license plates.

National University of Singapore

Research Intern with Prof. Mohan Kankanhalli
Singapore · Sep 2017 – May 2018

Worked on multimodal (EEG, and eye tracking) affect recognition from facial expression stimuli at the SeSaMe (Sensor Enhanced Social Media) Centre. Published at ICMI 2018.

Education
	University of Pittsburgh PhD in Intelligent Systems Pittsburgh, United States · 2019 – 2025

	IIIT Hyderabad MS by Research in Electronics and Communication · 8.5 / 10.00 Hyderabad, India · 2015 – 2018 Thesis: “Gender Differences in Facial Emotion Perception for User Profiling via Implicit Behavioral Signals” Advised by Prof. Ramanathan Subramanian at the Center for Visual Information Technology
	VIT Vellore Bachelors in Electronics and Communication · 8.98 / 10.00 Hyderabad, India · 2011 – 2015