Problem
Research questions and friction points this paper is trying to address.
Video Object-Centric Learning
Decoder
Spatial Anchoring
Semantic Information
Semantic Slots for Video Object-Centric Learning
🤖 AI Summary
该论文针对视频对象中心学习中解码器的局限性，提出使用基于Transformer的SemanticSlots方法，提高对象位置不变性和特征匹配能力，显著提升了性能。
📝 Abstract
Video Object-Centric Learning (OCL) has traditionally focused on refining the encoder architecture to ensure temporal consistency. In this paper, we argue that the primary bottleneck lies in the decoder. We show that traditional decoders force slots to be spatially anchored, hindering their ability to adapt to motion. We propose SemanticSlots, which uses a Transformer-based decoder that leverages image context, relieving slots from encoding boundary precision and spatial location. This allows slots to function as semantic queries that are inherently object position invariant, retrieving matching features rather than memorizing coordinates. More importantly, this property allows slots computed from a single frame to decompose subsequent video frames, eliminating the need for complex temporal predictors or auxiliary temporal losses. Results on YouTube-VIS show that SemanticSlots improves upon VideoSAUR by 31 points in mBO and outperforms current state-of-the-art methods by 21 points, achieving 86.6% ARI and 62.8% mBO.
Innovation
Methods, ideas, or system contributions that make the work stand out.
SemanticSlots
Transformer-based decoder
image context
object position invariance
temporal consistency
🔎 Similar Papers
No similar papers found.
