{
  "tool": "hotel-lobby-ai",
  "date": "2026-10-10",
  "design": "One video request per condition. The left cat reference differs; the right performer, prompt, motion reference and video settings are held fixed. Independently generated images differ in anatomy and pose, so this is an illustrative comparison, not an isolated-pose benchmark.",
  "audio": "Both pet outputs are original silent generation files. The tool adds the template soundtrack in its separate export step.",
  "conditions": [
    {
      "condition": "quadruped",
      "inputImage": {
        "taskId": "task_01a1250c3ff772fab387e05ae0eefd57",
        "request": {
          "model": "gpt-image-2",
          "prompt": "Create a clean photorealistic reference photograph. Exactly ONE fictional orange tabby cat with vivid orange striped fur, white chin and chest, green eyes, and a plain black sleeveless fabric vest fitted to the torso. Single photograph and viewpoint. Show the entire body, ears, all paws and tail within the frame, with comfortable margins. Neutral warm pale-beige seamless studio background, soft even light, sharp facial features. No human, no other animal, no duplicate, no inset portrait, no collage, no mirror, no microphone, no text, no watermark or logo. Portrait 3:4 composition. The cat is a natural quadruped standing on all FOUR paws, body horizontal, all four legs clearly separated and visible. Cat anatomy and a normal feline body; do not make it human-shaped. Face turned toward the camera, slight three-quarter body view.",
          "size": "3:4",
          "resolution": "2k",
          "n": 1
        },
        "status": "completed",
        "output": {
          "image_urls": [
            "https://file.deepytb.com/reapi/tasks/task_01a1250c3ff772fab387e05ae0eefd57/0.png"
          ]
        }
      },
      "video": {
        "taskId": "task_01a1250cf213760e8d6c491a795b0974",
        "request": {
          "model": "doubao-seedance-2.5-face",
          "prompt": "Create a single continuous two-person rap performance. Image 1 is the LEFT performer; Image 2 is the RIGHT performer. Preserve each character’s face, hair, clothing and identity from their own image throughout the shot. Use Video 1 only for choreography, body gestures, mouth movements, camera framing and performance timing. Do not copy appearance or line-art styling from Video 1. Both performers share one silver microphone against a warm orange studio backdrop. Follow the reference’s turns, arm movements and exchanges while keeping each character on their assigned side. Natural coherent anatomy, consistent hands, expressive faces. No extra people, text, cuts or scene changes. Render the characters in the visual style of their uploaded images.",
          "image_urls": [
            "https://file.deepytb.com/reapi/tasks/task_01a1250c3ff772fab387e05ae0eefd57/0.png",
            "https://file.deepytb.com/reapi/tasks/task_01a1246545c3712e9f9fea0176408e50/0.png"
          ],
          "video_urls": [
            "https://file.deepytb.com/tools/hotel-lobby-ai/motion-reference-v1.mp4"
          ],
          "duration": 15,
          "resolution": "480p",
          "size": "16:9",
          "generate_audio": false,
          "omni_reference_task_type": "reference",
          "draft": true,
          "content_filter": true
        },
        "status": "completed",
        "output": {
          "video_urls": [
            "https://file.deepytb.com/reapi/tasks/task_01a1250cf213760e8d6c491a795b0974/0.mp4"
          ]
        },
        "submitCount": 1,
        "startedAt": "2026-10-10T09:02:41.173244+00:00",
        "completedAt": "2026-10-10T09:06:07.238441+00:00"
      }
    },
    {
      "condition": "upright",
      "inputImage": {
        "taskId": "task_01a1250c4666777c9653ea1174e59634",
        "request": {
          "model": "gpt-image-2",
          "prompt": "Create a clean photorealistic reference photograph. Exactly ONE fictional orange tabby cat with vivid orange striped fur, white chin and chest, green eyes, and a plain black sleeveless fabric vest fitted to the torso. Single photograph and viewpoint. Show the entire body, ears, all paws and tail within the frame, with comfortable margins. Neutral warm pale-beige seamless studio background, soft even light, sharp facial features. No human, no other animal, no duplicate, no inset portrait, no collage, no mirror, no microphone, no text, no watermark or logo. Portrait 3:4 composition. The cat is an anthropomorphic upright performer standing on its TWO hind paws, body vertical and facing the camera, TWO front paws relaxed separately at its sides. Keep the same feline face, orange fur, stripes, white chin and chest, green eyes, black sleeveless vest and cat tail. No human hands; retain feline paws. Exactly four limbs total.",
          "size": "3:4",
          "resolution": "2k",
          "n": 1
        },
        "status": "completed",
        "output": {
          "image_urls": [
            "https://file.deepytb.com/reapi/tasks/task_01a1250c4666777c9653ea1174e59634/0.png"
          ]
        }
      },
      "video": {
        "taskId": "task_01a1250d0e43709b8433d4c50ea5c91b",
        "request": {
          "model": "doubao-seedance-2.5-face",
          "prompt": "Create a single continuous two-person rap performance. Image 1 is the LEFT performer; Image 2 is the RIGHT performer. Preserve each character’s face, hair, clothing and identity from their own image throughout the shot. Use Video 1 only for choreography, body gestures, mouth movements, camera framing and performance timing. Do not copy appearance or line-art styling from Video 1. Both performers share one silver microphone against a warm orange studio backdrop. Follow the reference’s turns, arm movements and exchanges while keeping each character on their assigned side. Natural coherent anatomy, consistent hands, expressive faces. No extra people, text, cuts or scene changes. Render the characters in the visual style of their uploaded images.",
          "image_urls": [
            "https://file.deepytb.com/reapi/tasks/task_01a1250c4666777c9653ea1174e59634/0.png",
            "https://file.deepytb.com/reapi/tasks/task_01a1246545c3712e9f9fea0176408e50/0.png"
          ],
          "video_urls": [
            "https://file.deepytb.com/tools/hotel-lobby-ai/motion-reference-v1.mp4"
          ],
          "duration": 15,
          "resolution": "480p",
          "size": "16:9",
          "generate_audio": false,
          "omni_reference_task_type": "reference",
          "draft": true,
          "content_filter": true
        },
        "status": "completed",
        "output": {
          "video_urls": [
            "https://file.deepytb.com/reapi/tasks/task_01a1250d0e43709b8433d4c50ea5c91b/0.mp4"
          ]
        },
        "submitCount": 1,
        "startedAt": "2026-10-10T09:02:49.508406+00:00",
        "completedAt": "2026-10-10T09:05:47.613884+00:00"
      }
    }
  ],
  "observations": {
    "method": "Reference images inspected; video review used frames at 0, 5, 10 and 14.5 seconds plus a one-frame-per-second sequence. This is frame-sequence review, not continuous playback or quantitative timing analysis.",
    "shared": [
      "Both sampled outputs show an upright orange cat on the left and the same adult woman on the right",
      "Black cat vest and ivory/white female outfit remain recognizable across sampled times",
      "Orange studio and hanging silver microphone remain visible",
      "Crop excludes lower legs; complete limb-count and hind-paw placement cannot be assessed",
      "No additional performer visible in sampled frames; not a frame-by-frame guarantee"
    ],
    "quadruped": [
      "First sampled frame already has upright torso, not original horizontal quadruped body",
      "White-tipped front paws form gestures; by approximately 5/14.5 seconds, extended digit-like shapes and pink pads visible",
      "Cat forelimb position changes across sampled sequence; no measured exact match to reference"
    ],
    "upright": [
      "Upright torso preserved in sampled frames",
      "Raised forelimb produces pointing and thumb-like hand shapes, rather than consistently natural feline paws",
      "Cat points through several early/middle sampled frames; no quantitative timing comparison"
    ],
    "interpretation": "An upright preparatory input was not necessary for this one quadruped-input output to show an upright upper body. Both outputs anthropomorphized paws. No universal requirement or causal posture effect established. No overall quality ranking.",
    "limitations": [
      "One output per condition",
      "Independently generated cats",
      "Human-oriented hair/hands prompt unchanged",
      "No seed fixed",
      "No audio-quality/lip-sync scoring",
      "No full-body hind-limb assessment",
      "Frame-sequence review only, not continuous playback"
    ],
    "audio": "Both probed outputs have one h264 video stream and no audio stream; 854x480, 24fps, 15.041667s"
  }
}