과도한 사회적 상호작용과 불필요한 콘텐츠가 혼재된 기존 SNS의 한계를 극복하고자 풍경과 동물 사진에 특화된 새로운 플랫폼을 제안하며, 효율적인 콘텐츠 분류를 위해 VGG-16, Inception, ResNet50, EfficientNet-B0 등 CNN기반 전이학습 모델들의 성능을 비교 분석한다.
This study proposes a new platform specialized in landscape and animal photography to overcome the limitations of existing SNS platforms, which are characterized by excessive social interactions and unnecessary mixed content. The performance of CNN-based transfer learning models including VGG-16, Inception, ResNet50 and EfficientNet is compared and analyzed for efficient content classification.