Download MBench-T/samples/object/vobj_sample_140_c141d12b/sample.json from studyOverflow/MBench-Data: direct link, hf CLI and curl.
- Browser
- Download file 1.51 kB
-
https://huggingface.co/datasets/studyOverflow/MBench-Data/resolve/main/MBench-T/samples/object/vobj_sample_140_c141d12b/sample.json
- Command line
-
hf download hf://datasets/studyOverflow/MBench-Data/MBench-T/samples/object/vobj_sample_140_c141d12b/sample.json
-
curl -L -o sample.json https://huggingface.co/datasets/studyOverflow/MBench-Data/resolve/main/MBench-T/samples/object/vobj_sample_140_c141d12b/sample.json
1.51 kB
| { | |
| "sample_id": "vobj_sample_140_c141d12b", | |
| "subset": "object", | |
| "caption_segments": [ | |
| "A red rectangular container is clearly visible in the foreground, stacked among others on the deck of the cargo ship ULTIMA, floating on calm ocean waters.", | |
| "The camera begins to pan left, and the red rectangular container remains visible as it moves toward the edge of the frame.", | |
| "The camera continues panning left, and the object is now fully out of frame, no longer visible in the scene.", | |
| "The camera starts to pan right, returning toward the original area where the container was initially seen.", | |
| "As the camera completes its rightward pan, the same red rectangular container is clearly visible again in its original position on the ship's deck, as the camera continues to pan right." | |
| ], | |
| "object_card": { | |
| "object_name": "container", | |
| "core_attributes": [ | |
| "red", | |
| "rectangular", | |
| "stacked" | |
| ], | |
| "scene_context": [ | |
| "cargo ship", | |
| "ocean", | |
| "shipping container stack" | |
| ], | |
| "must_keep": [ | |
| "focus on the container", | |
| "identify true color from original image" | |
| ], | |
| "must_avoid": [ | |
| "mentioning the mask", | |
| "describing the mask color" | |
| ], | |
| "visibility_plan": { | |
| "seg1": "clear full view", | |
| "seg2": "still visible while camera starts moving away", | |
| "seg3": "fully out of frame", | |
| "seg4": "camera returns toward the original area", | |
| "seg5": "same object clearly visible again" | |
| } | |
| } | |
| } | |