Dependable emotion generation is an essential requirement for enhancing affective computing studies, especially for the development and assessment of strong machine learning models. This study tackles the practical issue of finding suitable, already- existing video stimuli that reliably trigger particular emotional responses. Our main aim was to confirm a selected collection of video clips that can evoke the six fundamental emotions (happiness, sadness, anger, fear, disgust, surprise) along with a neutral condition. In this research, 24 individuals from Vadodara, Gujarat, India, watched the chosen video clips. Emotional reactions were assessed via self-report ratings on specific emotions and a 5-point scale for valence-arousal, and confirmed by physiological indicators (GSR and PPG). The findings reveal an overall emotion elicitation accuracy of 77.38% & 69.64% according to self-reports, and a Cohen’s Kappa value of 0.736 & 0.645 respectively for Session-1 & Session-2, signifying significant alignment between intended and perceived emotions. Valence-arousal mapping validated anticipated emotional distributions. Additionally, substantial differences (p