将大模型能力蒸馏到轻量级模型，实现设备端视频概念分割
EfficientSAM3: Progressive Hierarchical Distillation for Video Concept Segmentation from SAM1, 2, and 3
- 分三阶段渐进式蒸馏，逐步压缩大模型特征
- 在VOS数据集上达到高精度与低延迟平衡
- 适合移动端实时视频分割应用
Segment Anything Model 3（SAM3）通过提示驱动的概念分割（PCS）提升了图像和视频的视觉理解能力，但其统一架构（共享视觉主干、DETR风格检测器、密集记忆追踪器）难以在设备端部署。本文提出EfficientSAM3，一种基于渐进分层蒸馏（PHD）的轻量级模型系列，将SAM3的能力逐步迁移至小型学生模型：（1）编码器蒸馏通过在SA-1B上使用提示回路训练对齐图像特征；（2）时序记忆蒸馏用基于Perceiver的紧凑模块替代密集记忆，高效压缩与检索时空特征；（3）端到端微调在官方SAM3 PCS数据上优化完整流程，保留概念级性能。PHD生成多种变体，采用RepViT、TinyViT和EfficientViT主干，实现设备端概念分割与追踪，同时保持对教师行为的高保真度。我们在多个主流VOS数据集上进行基准测试，并与多种相关方法对比，展现出优异的性能-效率权衡。
原文摘要 · Abstract (English)
The Segment Anything Model 3 (SAM3) advances visual understanding with Promptable Concept Segmentation (PCS) across images and videos, but its unified architecture (shared vision backbone, DETR-style detector, dense-memory tracker) remains prohibitive for on-device use. We present EfficientSAM3, a family of efficient models built on Progressive Hierarchical Distillation (PHD) that transfers capability from SAM3 to lightweight students in three stages: (1) Encoder Distillation aligns image features via prompt-in-the-loop training on SA-1B; (2) Temporal Memory Distillation replaces dense memory with a compact Perceiver-based module trained on SA-V to compress and retrieve spatiotemporal features efficiently; and (3) End-to-End Fine-Tuning refines the full pipeline on the official SAM3 PCS data to preserve concept-level performance. PHD yields a spectrum of student variants using RepViT, TinyViT, and EfficientViT backbones, enabling on-device concept segmentation and tracking while maintaining high fidelity to teacher behavior. We benchmark on popular VOS datasets, and compare with varies of releated work, achieing strong performance-efficiency trade-offs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。