SlotCurri

Reconstruction-Guided Slot Curriculum:
Addressing Object Over-Fragmentation in Video Object-Centric Learning

CVPR 2026

WonJun Moon, Hyun Seok Seong, Jae-Pil Heo

Sungkyunkwan University

https://arxiv.org/abs/2603.22758

---

Abstract

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.

Key Contributions

Component	Description
Slot Curriculum	Progressively allocates slots where reconstruction error is high
Structure-Aware Loss	SSIM-based loss to sharpen semantic boundaries for precise slot partitioning
Cyclic Inference	Forward + backward temporal rolling for consistent slot representations

Code

https://github.com/wjun0830/SlotCurri

Qualitative Results on YouTube-VIS 2021

Qualitative Results on MOVi-C

---

Citation