Aniruddha Bala

aniruddha.b@samsung.com

I am a Generative AI Researcher in the Vision Intelligence team at Samsung, where I focus on instruction-based image editing and advancing practical applications of diffusion models. My work contributes to building secure, controllable, and user-friendly AI-powered editing solutions. Previously, I was part of Samsung’s Multimodal Intelligence team, where I developed on-device solutions such as Personal Knowledge Graph for cross-app data linking, automated Gallery Story Title Generation, and Screen AI for content understanding.

I earned my Master’s in Computational and Data Sciences from IISc Bangalore, under the guidance of Prof. Partha Talukdar, where my research focused on Open Knowledge Graph Embeddings. Before that, I completed my B.Tech in Electrical and Electronics Engineering at NIT Warangal.

[Google Scholar] [CV] [Linkedin] [X]

Recent News

[Jul 24] DCT-Shield has been selected as an ICCV 2025 Highlight 🌟
[Jun 25] Our work on DCT-Shield: A Robust Frequency Domain Defense against Malicious Image Editing accepted at ICCV 2025!! 🎉

Publications

Preprints
	Vid-Freeze: Protecting Images from Malicious Image-to-Video Generation Via Temporal Freezing Rohit Chowdhury, Aniruddha Bala, Siddharth Roheda, Rohit Chowdhury ICASSP 2026 (Under Review) [paper]
	Node-VNN: A Piecewise Continuous Volterra Neural Network Siddharth Roheda, Aniruddha Bala, Rohit Chowdhury, Rohan Jaiswal ICASSP 2026 (Under Review) [paper]
	GalaxyEdit: Large-Scale Image Editing Dataset with Enhanced Diffusion Adapter Rohan Jaiswal, Aniruddha Bala, Siddharth Roheda, Rohit Chowdhury ICASSP 2026 (Under Review) [paper]
	CART: Compositional AutoRegressive Transformer for Image Generation Siddharth Roheda, Rohit Chowdhury, Aniruddha Bala, Rohan Jaiswal AAAI 2026 (Under Review) [paper]
2025
	DCT-Shield: A Robust Frequency Domain Defense against Malicious Image Editing Aniruddha Bala, Rohit Chowdhury, Rohan Jaiswal, Siddharth Roheda ICCV 2025 (Highlight) [project page] [paper] [code]
2023
	Multi-Hop Relation Aware Representations for Inductive Knowledge Graphs Aniruddha Bala, Ankit Sharma, Shlok Sharma, Pinaki Bhaskar ICON 2023 [paper]
2021
	An On-Device Deep Learning Approach for Attribute Extraction Mahesh Gorijala, Aniruddha Bala, Pinaki Bhaskar, Krishnaditya, Vikram Mupparthi ICON 2021 [paper]
2020
	Decomposing camera and object motion for an improved video sequence prediction Meenakshi Sarkar, Debasish Ghose, Aniruddha Bala Workshop on Pre-Registration in Machine Learning, NeurIPS 2020 [project page] [paper] [code]

Honors

Received Samsung Best Paper Award 2025 for our work DCT-Shield.

This website template is taken from here