videox_fun/video_caption/datasets/panda_70m/frame_caption.jsonl

{"video_path":"--C66yU3LjM_2-Scene-001.mp4","image_caption_model":"llava-v1.6-vicuna-7b","prompt":"Please describe this image in detail.","sampled_frame_idx":[119],"sampled_frame_caption":[" The image captures a moment of creativity in progress. A person's hands are meticulously crafting a large card, which is predominantly covered in a glittery paper. The vibrant green and white plaid shirt they are wearing adds a touch of color to the scene. \n\nThe card is not alone in this image - it's resting on a white counter draped with a white and silver striped cloth. On the same counter, there are other objects, such as two books, a round silver tray, and a striped fabric gently holding the card. \n\nInterestingly, the counter isn't the only surface in the image. A white curtain can also be seen, adding depth to the scene. The entire setting gives off an atmosphere of creativity and careful craftsmanship."]}
{"video_path":"--C66yU3LjM_2-Scene-002.mp4","image_caption_model":"llava-v1.6-vicuna-7b","prompt":"Please describe this image in detail.","sampled_frame_idx":[119],"sampled_frame_caption":[" This image depicts a close-up view of a person's hands engaging with fabric bow-making supplies. The person is wearing a green and white checkered shirt and bracelets. The focus is on the hands and the materials being handled, which include a patterned bow and what appears to be a ribbon or bandaged material with a pearlescent texture. In the background, there is a blurred view of a room setting with a vase and a decorative item on a countertop. The style of the image is candid and appears to be taken in a domestic setting. There are no visible texts in the image."]}
{"video_path":"--C66yU3LjM_2-Scene-003.mp4","image_caption_model":"llava-v1.6-vicuna-7b","prompt":"Please describe this image in detail.","sampled_frame_idx":[119],"sampled_frame_caption":[" The image shows a close-up of a person's hands, focusing on a detailed textile that appears to be a decorative item with multiple layers. This object resembles a fan, with a metallic, reflective material at the center. The hands, visible in focus, are holding up the fabric while pointing to a small, circular object next to it. This circular object has a metallic base and is decorated with nacreous or iridescent speckles, possibly in shades of gold and silver.\n\nIn the background, there is a corner of a room with a window covered by sheer white curtains. The person is wearing a watch that is visible on their left wrist, and their attire consists of a green patterned shirt. There are no visible texts within the image. The style of the photograph is candid and seems to be taken to document a specific items the person is examining or showcasing."]}
{"video_path":"--C66yU3LjM_2-Scene-004.mp4","image_caption_model":"llava-v1.6-vicuna-7b","prompt":"Please describe this image in detail.","sampled_frame_idx":[44],"sampled_frame_caption":[" The image is a photograph taken at an angle showing an individual with their right hand holding what appears to be a wool hat with a small crumpled top, possibly made of felt. The hat has a glittery thread or material scattered across its surface. The individual is wearing a green and black checkered patterned shirt and is holding the hat close to their chest, displaying it prominently. They have a watch on their left wrist and are wearing a ring on the same finger, suggesting they might be engaged or married. The room has a warm tone, with a lamp next to the person casting a glow on the scene. In the background, there's a partial view of a wall with items on it to the right, but the details are not clear. The image does not contain any visible text."]}
{"video_path":"--C66yU3LjM_2-Scene-005.mp4","image_caption_model":"llava-v1.6-vicuna-7b","prompt":"Please describe this image in detail.","sampled_frame_idx":[54],"sampled_frame_caption":[" In the image, a person is engaged in a crafting activity. They are using a craft tool, specifically a stump carving knife and a bag pulling device, to create a craft item. The person is wearing a green and white plaid shirt. The craft item being made is a bow, which the person is holding in their hands. \n\nThe scene is set in a room with a white countertop and a white lamp stand. There is also a wooden bowl on the countertop. On the countertop, a pair of scissors can be seen lying next to the person's work area. The focus of the image is on the person and their crafting activity, with all other objects in the image serving as accessories to this central action. \n\nThe lighting in the room is soft and ambient, suggesting an indoor setting and a calm atmosphere conducive to creative work. The overall impression is of someone deeply involved in their craft, demonstrating mid-process as they work on their bow."]}