SAM3Dual: A 3rd Place Solution to the MOSEv2 Track, 8th LSVOS Challenge

2026-08-22

AI Summary

本文提出了SAM3Dual，通过重组时间记忆在推理阶段提高长期视频对象分割性能，无需额外训练，实现了64.37的J&F得分，在MOSEv2赛道中排名第三。

Abstract

We present SAM3Dual, our third-place solution to the MOSEv2 track of the 8th Large-scale Video Object Segmentation (LSVOS) Challenge at ECCV 2026. SAM3Dual is a training-free inference extension of pretrained SAM 3 that explicitly separates temporal memory into a short-term branch for recent observations and a long-term branch for interval-sampled historical representations. The two memory responses are combined using a deterministic sequence-relative fusion schedule and conservatively modulated by the previous-frame object confidence. All pretrained SAM 3 parameters remain frozen, requiring no task-specific training, fine-tuning, test-time training, or online parameter optimization. The complete system achieved an official J&F score of 64.37 and ranked third in the MOSEv2 track. This result highlights the potential of reorganizing temporal memory entirely at inference time to obtain competitive long-term VOS performance while preserving the pretrained model.

Problem

• Large-scale Video Object Segmentation
• Temporal Memory
• Pretrained Model
• Inference Time

Innovation

• training-free inference
• temporal memory separation
• sequence-relative fusion schedule
• long-term VOS performance