3d-estimation computer-vision depth-estimation foundation-models mllm multimodal-llm point-clouds robotics scene-understanding spatial-intelligence stereo-vision transformer vision-language-model vision-transformer zero-shot-learning

1-hour-video-language-understanding benchmark-dataset egocentric-videos evals gemini-pro gpt-4 long-context-understanding long-form-video-language-understanding multimodal-large-language-models multiple-choice-questions navigation neurips-2024 perception reasoning spatial-intelligence summarization video-language-understanding visual-reasoning