Gallileo-4D：用于动态4D重建的冻结主干集成模型
Gallileo-4D: Frozen Backbone Ensemble for Dynamic 4D Reconstruction
- OdaxAI Research（OdaxAI研究院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究为PhysAI动态4D重建挑战赛设计了冻结主干集成模型，通过融合三种解码配置实现零训练成本下的APD提升，获27支队伍中第三名。
AI中文摘要：
本文介绍了我们参加PhysAI动态4D重建挑战赛的参赛方案，该方案在最终排行榜上以0.58356的APD指标在27支队伍中排名第三，且未进行任何梯度更新。最初的计划并非如此：对预训练的4D主干网络进行13种微调配置，其中12种降低了挑战赛得分，而这12种中的11种同时提升了局部验证性能。我们将这种性能反转归因于基准的结构：评估集中仅有25%属于训练所用的数据变体，因此适配可用数据的更新会损害剩余75%数据所依赖的预训练特征。因此，我们的系统冻结了主干网络，将预算用于推理阶段，通过凸加权融合三种解码配置——时间步长为3、水平翻转测试时数据增强、以及密集步长为1——的集成模型，该集成模型在零训练成本下，相较于冻结基线模型恢复了+0.041的APD提升，超过了任何训练运行所达到的效果。
英文摘要：
We describe our entry to the PhysAI Dynamic 4D Reconstruction Challenge, which placed third of 27 teams at 0.58356 APD on the final leaderboard, without a single gradient update. This was not the plan: of thirteen fine-tuning configurations of a pre-trained 4D backbone, twelve degraded the challenge score, and eleven of those twelve improved local validation at the same time. We trace this inversion to the structure of the benchmark: only 25% of the evaluation set belongs to the data variant released for training, so updates that fit the available data damage the pre-trained features the remaining 75% relies on. Our system therefore freezes the backbone and spends its budget at inference time, fusing three decoding configurations -- temporal stride-3, horizontal-flip test-time augmentation, and dense stride-1 -- under a convex weighting. The ensemble recovers +0.041 APD over the frozen baseline, more than any training run achieved, at zero training cost.
