重建引导的槽课程：解决视频目标中心学习中的目标过分割问题
Reconstruction-Guided Slot Curriculum: Addressing Object Over-Fragmentation in Video Object-Centric Learning
针对槽注意力模型在视频目标中心学习中普遍存在的过分割问题，提出按重建误差逐步扩展槽容量的课程式训练策略，并配合结构感知损失与循环推理提升时序一致性；这套「按需分配表征容量」的思路可直接迁移到任何使用固定槽/查询数量的视频或图像密集预测模型（如 DETR 类查询、视频实例分割）。
原文依据 Training starts with only a few coarse slots and progressively allocates new slots where reconstruction error remains high
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视频目标中心学习旨在将原始视频分解为少量目标槽，但现有的槽注意力模型常常遭受严重的过分割问题。这是因为模型被隐式地鼓励占满所有槽以最小化重建目标，从而用多个冗余槽表示单个目标。我们通过重建引导的槽课程（SlotCurri）来解决这一局限。训练从少量粗粒度槽开始，并在重建误差仍然较高的位置逐步分配新槽，从而只在需要的地方扩展容量，从一开始就防止碎片化。然而，在槽扩展过程中，只有当粗粒度语义已经良好分离时，有意义的子部件才能出现；而在初始槽预算较小且使用 MSE 目标的情况下，语义边界仍然模糊。因此，我们在 MSE 之外增加了一个结构感知损失，用于保留局部对比度和边缘信息，以促使每个槽锐化其语义边界。最后，我们提出了循环推理，使槽沿帧序列向前再向后滚动，即使在最早的帧中也能产生时序一致的目标表示。综合起来，SlotCurri 通过在重建失败处分配表征容量来解决目标过分割问题，并进一步由结构线索和循环推理增强。在 YouTube-VIS 上 +6.8、在 MOVi-C 上 +8.3 的显著 FG-ARI 提升验证了 SlotCurri 的有效性。我们的代码可在 github.com/wjun0830/SlotCurri 获取。
英文原文摘要 1577 字符
Video Object-Centric Learning seeks to decompose raw videos into a small set of object slots, but existing slot-attention models often suffer from severe over-fragmentation. This is because the model is implicitly encouraged to occupy all slots to minimize the reconstruction objective, thereby representing a single object with multiple redundant slots. We tackle this limitation with a reconstruction-guided slot curriculum (SlotCurri). Training starts with only a few coarse slots and progressively allocates new slots where reconstruction error remains high, thus expanding capacity only where it is needed and preventing fragmentation from the outset. Yet, during slot expansion, meaningful sub-parts can emerge only if coarse-level semantics are already well separated; however, with a small initial slot budget and an MSE objective, semantic boundaries remain blurry. Therefore, we augment MSE with a structure-aware loss that preserves local contrast and edge information to encourage each slot to sharpen its semantic boundaries. Lastly, we propose a cyclic inference that rolls slots forward and then backward through the frame sequence, producing temporally consistent object representations even in the earliest frames. All combined, SlotCurri addresses object over-fragmentation by allocating representational capacity where reconstruction fails, further enhanced by structural cues and cyclic inference. Notable FG-ARI gains of +6.8 on YouTube-VIS and +8.3 on MOVi-C validate the effectiveness of SlotCurri. Our code is available at github.com/wjun0830/SlotCurri.