Multimodal research in vision and language: A review of current and emerging trends

Deep Learning and its applications have cascaded impactful research development with a diverse range of modalities present in the real-world data. More recently, this has enhanced interests intersection Vision Language arena numerous fast-paced growth. In paper, we detailed overview latest trends pertaining to visual language modalities. We look at their task formulations how solve various problems related semantic perception content generation. also address task-specific trends, along evaluation strategies upcoming challenges. Moreover, shed some light on multi-disciplinary patterns insights that emerged recent past, directing field toward more modular transparent intelligent systems. This survey identifies key gravitating literature VisLang attempts unearth directions is heading toward.