Eyes on Poverty: A Comprehensive Review of Computer Vision and Textual Data Integration for Poverty Prediction in India

Poverty, a pervasive and multidimensional societal issue, continues challenging policymakers, researchers, and humanitarian organizations. In the quest to address this complex problem, leveraging advancements in data science and computer vision has emerged as a promising avenue. This review paper delves into the fusion of multimodal data— comprising images and text—as a transformative approach for poverty prediction, focusing on India, a nation marked by its rich diversity and unique challenges. The paper elucidates poverty's multidimensional nature, emphasizing its economic, social, and spatial dimensions. Traditional poverty assessment methods are scrutinized, revealing their inherent limitations, particularly in a dynamic and rapidly evolving nation like India. Recognizing the limitations of these conventional approaches, the paper sets the stage for exploring the novel paradigm of multimodal data analysis. The article surveys prominent datasets employed in the field, illustrating how these information repositories empower researchers to build predictive models capable of dissecting the intricate nuances of poverty. The core of this review revolves around a comprehensive evaluation of cutting-edge methodologies and techniques underpinning poverty prediction, particularly in the Indian context, revealing how they unlock insights into housing conditions, agricultural productivity, healthcare access, and socio-economic disparities. This review paper transcends the boundaries of traditional poverty assessment methodologies by venturing into the realm of multimodal data analysis. By weaving together, the intricate threads of images and text, it unravels fresh insights into poverty dynamics, particularly within the unique socio-economic landscape of India. This synthesis of multidisciplinary knowledge serves as a guiding beacon for researchers, practitioners, and policymakers alike, beckoning them to harness the transformative power of data science in the relentless pursuit of poverty alleviation.