Semantic Slots for Video Object-Centric Learning

• URL: http://arxiv.org/abs/2608.21636v1
• Date: Fri, 21 Aug 2026 21:12:50 GMT
• Title: Semantic Slots for Video Object-Centric Learning
• Abstract summary: We show that traditional decoders force slots to be spatially anchored, hindering their ability to adapt to motion.<n>We propose SemanticSlots, which uses a Transformer-based decoder that leverages image context.<n>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.
• Score: 8.62968609670716
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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.

Related papers

• Internalizing Temporal Consistency in Video Object-Centric Learning without Explicit Regularization [34.98521123729049]
  Slot-Slot Contrastive (SSC) loss has become the cornerstone for state-of-the-art (SOTA) video Object-Centric Learning (OCL) methods.<n>We propose a paradigm shift: temporal consistency is better enforced as an implicit model design rather than an explicit loss.
  arXiv  Detail & Related papers  (2026-05-29T16:28:21Z)
• From Vicious to Virtuous Cycles: Synergistic Representation Learning for Unsupervised Video Object-Centric Learning [45.1920794546889]
  We introduce a virtuous cycle where the encoder and decoder mutually refine one another.<n>By bridging the representational gap between the encoder and decoder, SRL achieves state-of-the-art results on video object-centric learning benchmarks.
  arXiv  Detail & Related papers  (2026-02-03T11:11:58Z)
• FRAME: Pre-Training Video Feature Representations via Anticipation and Memory [55.046881477209695]
  FRAME is a self-supervised video frame encoder tailored for dense video understanding.<n>It learns to predict current and future DINO patch features from past and present RGB frames.<n>It consistently outperforms image encoders and existing self-supervised video models.
  arXiv  Detail & Related papers  (2025-06-05T19:44:47Z)
• A Spatial-Temporal Deformable Attention based Framework for Breast Lesion Detection in Videos [107.96514633713034]
  We propose a spatial-temporal deformable attention based framework, named STNet. Our STNet introduces a spatial-temporal deformable attention module to perform local spatial-temporal feature fusion. Experiments on the public breast lesion ultrasound video dataset show that our STNet obtains a state-of-the-art detection performance.
  arXiv  Detail & Related papers  (2023-09-09T07:00:10Z)
• Distortion-Aware Network Pruning and Feature Reuse for Real-time Video Segmentation [49.17930380106643]
  We propose a novel framework to speed up any architecture with skip-connections for real-time vision tasks. Specifically, at the arrival of each frame, we transform the features from the previous frame to reuse them at specific spatial bins. We then perform partial computation of the backbone network on the regions of the current frame that captures temporal differences between the current and previous frame.
  arXiv  Detail & Related papers  (2022-06-20T07:20:02Z)
• Deep Multi-Branch Aggregation Network for Real-Time Semantic Segmentation in Street Scenes [32.54045305607654]
  Many state-of-the-art real-time semantic segmentation methods tend to sacrifice some spatial details or contextual information for fast inference. We propose a novel Deep Multi-branch Aggregation Network (called DMA-Net) based on the encoder-decoder structure to perform real-time semantic segmentation in street scenes. Our proposed DMA-Net respectively obtains 77.0% and 73.6% mean Intersection over Union (mIoU) at the inference speed of 46.7 FPS and 119.8 FPS by only using a single NVIDIA GTX 1080Ti GPU.
  arXiv  Detail & Related papers  (2022-03-08T12:07:32Z)
• Stage-Aware Feature Alignment Network for Real-Time Semantic Segmentation of Street Scenes [59.81228011432776]
  We present a novel Stage-aware Feature Alignment Network (SFANet) for real-time semantic segmentation of street scenes. By taking into account the unique role of each stage in the decoder, a novel stage-aware Feature Enhancement Block (FEB) is designed to enhance spatial details and contextual information of feature maps from the encoder. Experimental results show that the proposed SFANet exhibits a good balance between accuracy and speed for real-time semantic segmentation of street scenes.
  arXiv  Detail & Related papers  (2022-03-08T11:46:41Z)
• Collaborative Spatial-Temporal Modeling for Language-Queried Video Actor Segmentation [90.74732705236336]
  Language-queried video actor segmentation aims to predict the pixel-mask of the actor which performs the actions described by a natural language query in the target frames. We propose a collaborative spatial-temporal encoder-decoder framework which contains a 3D temporal encoder over the video clip to recognize the queried actions, and a 2D spatial encoder over the target frame to accurately segment the queried actors.
  arXiv  Detail & Related papers  (2021-05-14T13:27:53Z)
• Dense Interaction Learning for Video-based Person Re-identification [75.03200492219003]
  We propose a hybrid framework, Dense Interaction Learning (DenseIL), to tackle video-based person re-ID difficulties. DenseIL contains a CNN encoder and a Dense Interaction (DI) decoder. Our experiments consistently and significantly outperform all the state-of-the-art methods on multiple standard video-based re-ID datasets.
  arXiv  Detail & Related papers  (2021-03-16T12:22:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.