<![CDATA[The development of generative Artificial Intelligence (AI) technology has had a significant impact on the multimedia sector, particularly in image-to-video techniques that are capable of automatically transforming static images into videos. This study aims to analyze and compare the video quality produced by four AI platforms, namely Kling, Runway, PixVerse, and Pika, in the context of 2D animation. The method used is a comparative experimental approach combining quantitative and qualitative methods. The data consist of three rendered 2D animation images from Blender that were converted into 5-second videos using identical prompts on each platform. Quantitative evaluation was conducted through measurements of processing time, Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM). Meanwhile, qualitative evaluation involved panelists using a Likert scale to assess nine visual aspects. The results indicate that Pika and Runway excelled in processing time efficiency, with average times of 34.4 seconds and 36.3 seconds, respectively. Kling achieved the highest PSNR and SSIM values, with an average PSNR of 14.62 dB and an SSIM of 0.41, indicating the best technical quality. On the other hand, Runway received the highest ratings in terms of visual and aesthetic aspects based on respondent evaluations. Overall, no single platform outperformed the others across all aspects of the study. Therefore, the selection of a platform should be adjusted according to user needs, whether in terms of efficiency, technical quality, or visual aesthetics. This study highlights the importance of an integrated evaluation approach to produce a more comprehensive assessment of video quality.]]>
