{"submission_id":"f220e171-53c6-4c30-920f-5c7783c0a27f","submission_revision_id":"310e7603-1289-40a2-80d3-081b433e2540","revision_version":6,"submission_status":"under_review","submitted_at":"2026-09-14T05:37:44.669359Z","submitter_name":"Ipek Oztas","submitter_email":"ipek_oztas@brown.edu","task":{"status":"submitted","version":6,"extra_fields":{"difficulty_self_test":{"model":"GPT-5.6 Sol","estimated_score":"<1%"}},"submitted_at":"2026-09-14T05:37:44.669359+00:00","industry_domain":"3D Animation & Interactive Media","task_short_name":"robotexpressive_step_wave_reproduction_01","operating_system":"MacOS","task_description":"You are a 3D artist using Blender. Rig and animate the provided RobotExpressive character to reproduce the body motion in the supplied reference video.\n\nStarting materials:\n\n* input/character.obj: neutral, unrigged character mesh.\n* input/character.mtl: character materials.\n* input/reference.mp4: an 8-second reference performance at 24 fps.\n\nThe performance includes an initial weight shift, a step toward the character’s anatomical left, raising the right arm, two waves, lowering the arm, and returning to the starting stance. Match the reference’s poses, timing, movement direction, and overall motion.\n\nRequirements:\n\n* Preserve all 14 supplied character mesh parts and their recognizable geometry and materials.\n* Create a functional armature that drives the character’s body animation. Rigid bone parenting or skinning is acceptable where appropriate for this segmented robot.\n* Use a skeleton appropriate to the supplied character. No particular bone names or bone count are required.\n* Focus on body motion. Finger animation, facial animation, and lip synchronization are not required.\n* Use a fixed camera that approximates the reference viewpoint and keeps the full performance visible.\n* Use a simple floor, background, and lighting. Exact pixel matching of lighting is not required.\n* Blender Python scripting is permitted.\n* Do not retrieve the original rig, original GLB, or existing animation from external sources. Work from the supplied files.\n\nDeliver:\n\n* output/submission/final.blend\n* output/submission/preview.mp4\n\nThe preview must be 960×960, 24 fps, 192 frames, and approximately 8 seconds long. The Blender project must contain the actual rig, materials, camera, and animation used to produce the preview, with necessary resources packed or included.\n\nThe evaluator will reopen the Blender project and replay the animation. A reference video embedded on a plane, copied preview, or animated image sequence does not satisfy the task.","software_licensing":"Free/Open Source","evaluation_criteria":null,"verification_method":"Validity gate:\nBoth output files must exist and be readable. The Blender scene must contain the supplied character and a functional animated armature that drives its motion. The evaluator must be able to replay and render the scene. Failure of this gate gives a score of zero.\n\nScore valid submissions using:\n\n* 50% motion reproduction: action sequence, anatomical laterality, two distinct waves, body poses, trajectory, and event timing relative to the reference.\n* 20% visual and structural fidelity: preservation of all character parts, coherent articulation, and absence of major new detachments or deformation failures.\n* 15% temporal/contact quality: foot placement during support phases, transitions, and absence of major unintended jitter or penetration.\n* 15% replay consistency: the evaluator-rendered scene agrees with the submitted preview.\n\nUse evaluator-rendered evidence as the authoritative representation of the submission. Do not award motion credit solely for a matching submitted video.\n\nCompare anatomical landmarks or corresponding mesh-part motion rather than requiring identical bone names, bone lengths, or rig implementation. Document the landmark mapping, coordinate alignment, numerical tolerances, and motion-event windows in the evaluation configuration.\n\nAllow plausible depth interpretations consistent with the single reference viewpoint. Do not require recovery of unobservable hidden 3D poses.\n\nKnown source-character features and documented imperfections in the reference, including floating-foot construction and occasional hand/shin contact, must not automatically disqualify a submission. Assess additional errors relative to the reference.\n\nValidate the evaluator using the reference project, an independently checked valid result, and deliberately incorrect submissions such as a static character, mirrored motion, missing wave, shifted timing, and copied preview with an incorrect scene.","software_and_version":"Blender 5.2.1 LTS","specific_requirements":null,"expected_completion_time":null},"files":[{"file_id":"1763e8e9-d7d1-4158-807a-88eaf6470492","filename":"reference.mp4","category":"input_materials","content_type":"video/mp4","size_bytes":1387760},{"file_id":"fdf9b3a3-40ec-4bb3-9546-2a6b100c0d3d","filename":"character.mtl","category":"input_materials","content_type":"application/octet-stream","size_bytes":587},{"file_id":"18694782-c5a4-4f28-b9b1-7f37738b7b79","filename":"character.obj","category":"input_materials","content_type":"model/obj","size_bytes":371405},{"file_id":"f96ceacc-25f4-4277-86ca-6e984e143fbd","filename":"stage_config.json","category":"reference_output","content_type":"application/json","size_bytes":3104},{"file_id":"669baa20-2d95-4a09-b6f0-7a7e29743c08","filename":"reference.blend","category":"reference_output","content_type":"application/octet-stream","size_bytes":411980},{"file_id":"122bd91d-73f7-4178-893f-e82f349ab4fd","filename":"ground_truth_motion.json","category":"reference_output","content_type":"application/json","size_bytes":2178973},{"file_id":"7f356c5c-c3c0-4113-8265-2ceee6356ba8","filename":"reference.mp4","category":"reference_output","content_type":"video/mp4","size_bytes":1387760},{"file_id":"d3b903f8-648f-4611-80f5-e4c0d2f42360","filename":"SOURCE_AND_LICENSE.md","category":"reference_output","content_type":"text/markdown","size_bytes":2930},{"file_id":"5309b761-ab2a-44d0-a804-781308444609","filename":"reference_metadata.json","category":"reference_output","content_type":"application/json","size_bytes":17380},{"file_id":"be1d445f-0329-4aeb-aa0b-49329e23e4ef","filename":"preview.mp4","category":"difficulty_self_test_evidence","content_type":"video/mp4","size_bytes":588258},{"file_id":"9e4f83a9-69da-4e56-bf8b-cb85a4f5713e","filename":"final.blend","category":"difficulty_self_test_evidence","content_type":"application/octet-stream","size_bytes":198207},{"file_id":"a279e71b-2584-4cce-92c6-9f8cb52c15a2","filename":"run_evaluator.py","category":"reference_output","content_type":"text/x-python-script","size_bytes":36250},{"file_id":"5f21cae7-23c2-4478-bba0-98f331c56b3d","filename":"evaluator_extract.py","category":"reference_output","content_type":"text/x-python-script","size_bytes":12533},{"file_id":"625f1be8-2219-47ff-9a75-3aefc2c7feaa","filename":"evaluator_replay.py","category":"reference_output","content_type":"text/x-python-script","size_bytes":2503},{"file_id":"ea19c3cd-c954-4c32-8a06-36d645ee1d51","filename":"evaluator_config.json","category":"reference_output","content_type":"application/json","size_bytes":9894},{"file_id":"cb2fdfa1-5fc2-4db5-87f9-52e16b924890","filename":"manual_rubric.md","category":"reference_output","content_type":"text/markdown","size_bytes":2474}],"deep_review":{"overall_verdict":"Accept","review_text":"**Overall Verdict:** Accept — the task, evaluator, references, and self-test evidence are complete and end-to-end reachable, with only an AI-assisted provenance screening note.\n\n**Completeness Check:**\n- Instruction: Complete — specifies the motion sequence, laterality, geometry/rig requirements, scene setup, output contract, and prohibited shortcuts.\n- Inputs: Complete — `character.obj`, `character.mtl`, and the reference video are present and readable; the OBJ contains 14 mesh objects and the declared three-material palette.\n- Toolchain: Complete — Blender 5.2.1 LTS on macOS is appropriate; scripting and FFmpeg provide a feasible production/evaluation path without credentials.\n- Reference output: Complete — `stage_config.json`, `reference.blend`, `ground_truth_motion.json`, reference video, metadata, and licensing documentation define the expected scene, motion, coordinates, and provenance.\n- Rubric: Complete — evaluator configuration and scripts define landmark correspondence, laterality, timing windows, contact handling, replay checks, validity gates, and manual review provisions.\n\n**File and Task Reachability:**\nThe attachment set is `reachable`: all manifest entries were downloaded with matching byte counts and safely inspected, including `FILES.json`; both video copies; `character.obj`; `character.mtl`; `stage_config.json`; `reference.blend`; `ground_truth_motion.json`; `SOURCE_AND_LICENSE.md`; `reference_metadata.json`; `preview.mp4`; `final.blend`; `run_evaluator.py`; `evaluator_extract.py`; `evaluator_replay.py`; `evaluator_config.json`; and `manual_rubric.md`. The OBJ/MTL and JSON/Markdown files parsed successfully. The MP4 containers are H.264 streams with 960×960 dimensions, 192 frames, and 8-second duration. The Blender payloads were decompressed and structurally inspected; the submitted scene contains 14 numbered mesh parts, an armature, an animation action, camera, floor, and lighting. I did not launch Blender or execute submitted scripts, and this environment lacked a video decoder, so fresh replay, actual evaluated deformation, and frame-by-frame visual quality remain unverified reviewer-tooling limitations rather than attachment access barriers. The task itself is `reachable`: a contributor can obtain the declared free toolchain, build the scene from the supplied inputs, produce both deliverables, and run the objective evaluator on one suitable machine. License keys, accounts, tokens, and other credentials are not required from the contributor. Confidential, proprietary, or internal material may be evaluated as submitted evidence without requiring additional disclosure.\n\n**Special Screening:**\n`ai_synthetic` — the reference metadata states that the reference animation was generated by headless Blender scripts and authored with AI assistance, and describes the action as newly authored without evidence of an authentic professional-performance origin. The source character is separately documented as a CC0 asset. The synthetic provenance should remain explicitly disclosed to evaluators.\n\n**Recommended Changes:**\nThe package is otherwise professionally specified and evaluator-ready; no implementation changes are required. Retain the explicit disclosure that the reference motion and ground truth are scripted AI-assisted synthetic material, and keep evaluator-only scene, ground-truth, and scoring assets separated from solver-visible inputs.","reviewed_at":"2026-09-14T05:49:53.959953Z"},"claim":null}