GPT-4o: Visual perception performance of multimodal large language models in piglet activity understanding

Benchmark Model Rank Results
mmr-total-on-mrr-benchmarkGPT-4o–Total Column Score: 457
zero-shot-video-question-answer-on-longvideobenchGPT-4o–Accuracy (% ): 64.0
zero-shot-video-question-answer-on-video-mmeGPT-4o–Accuracy (%): 77.2
zero-shot-video-question-answer-on-video-mmeGPT-4o mini–Accuracy (%): 68.9
zero-shot-video-question-answer-on-video-mme-w-o-subsGPT-4o–Accuracy (%): 70.3
zero-shot-video-question-answer-on-video-mme-w-o-subsGPT-4o mini–Accuracy (%): 62.3