14 papers · 1 filter
HSA: Hierarchical Slot Attention for Multi-granularity Scene-Decomposition
Neelu Madan, Rongzhen Zhao, Andreas Mogelmose +4
Slot attention is a powerful framework for object-centric learning, decomposing visual scenes into latent slots through iterative competitive attention. However, existing methods s…
Internalizing Temporal Consistency in Video Object-Centric Learning without Explicit Regularization
Rongzhen Zhao, Zhiyuan Li, Juho Kannala +1
Video Object-Centric Learning (OCL) aims to represent objects as \textit{slot} vectors and maintain their consistency across frames. Slot-Slot Contrastive (SSC) loss has become the…
Cycle Consistency in Video Object-Centric Learning
Rongzhen Zhao, Zhiyuan Li, Ruonan Wei +2
Self-supervised video Object-Centric Learning (OCL) aims to discover distinct objects and associate them across time, whereas self-supervised Multi-Object Tracking (MOT) focuses on…
Rethinking Temporal Consistency in Video Object-Centric Learning: From Prediction to Correspondence
Zhiyuan Li, Rongzhen Zhao, Wenyan Yang +3
The de facto approach in video object-centric learning maintains temporal consistency through learned dynamics modules that predict future object representations, called slots. We…
Object-Centric Vision Token Pruning for Vision Language Models
Guangyuan Li, Rongzhen Zhao, Jinhong Deng +2
In Vision Language Models (VLMs), vision tokens are quantity-heavy yet information-dispersed compared with language tokens, thus consume too much unnecessary computation. Pruning r…
Smoothing Slot Attention Iterations and Recurrences
Rongzhen Zhao, Wenyan Yang, Juho Kannala +1
Slot Attention (SA) lies at the heart of mainstream Object-Centric Learning (OCL). Image features can be aggregated into object-level representations by SA \textit{iteratively} ref…
Predicting Video Slot Attention Queries from Random Slot-Feature Pairs
Rongzhen Zhao, Jian Li, Juho Kannala +1
Unsupervised video Object-Centric Learning (OCL) is promising as it enables object-level scene representation and understanding as we humans do. Mainstream video OCL methods adopt…
Slot Attention with Re-Initialization and Self-Distillation
Rongzhen Zhao, Yi Zhao, Juho Kannala +1
Unlike popular solutions based on dense feature maps, Object-Centric Learning (OCL) represents visual scenes as sub-symbolic object-level feature vectors, termed slots, which are h…
MetaSlot: Break Through the Fixed Number of Slots in Object-Centric Learning
Hongjia Liu, Rongzhen Zhao, Haohan Chen +1
Learning object-level, structured representations is widely regarded as a key to better generalization in vision and underpins the design of next-generation Pre-trained Vision Mode…
Vector-Quantized Vision Foundation Models for Object-Centric Learning
Rongzhen Zhao, Vivienne Wang, Juho Kannala +1
Object-Centric Learning (OCL) aggregates image or video feature maps into object-level feature vectors, termed \textit{slots}. It's self-supervision of reconstructing the input fro…
Grouped Discrete Representation for Object-Centric Learning
Rongzhen Zhao, Vivienne Wang, Juho Kannala +1
Object-Centric Learning (OCL) aims to discover objects in images or videos by reconstructing the input. Representative methods achieve this by reconstructing the input as its Varia…
Multi-Scale Fusion for Object Representation
Rongzhen Zhao, Vivienne Wang, Juho Kannala +1
Representing images or videos as object-level feature vectors, rather than pixel-level feature maps, facilitates advanced visual tasks. Object-Centric Learning (OCL) primarily achi…
Organized Grouped Discrete Representation for Object-Centric Learning
Rongzhen Zhao, Vivienne Wang, Juho Kannala +1
Object-Centric Learning (OCL) represents dense image or video pixels as sparse object features. Representative methods utilize discrete representation composed of Variational Autoe…
Grouped Discrete Representation Guides Object-Centric Learning
Rongzhen Zhao, Vivienne Wang, Juho Kannala +1
Similar to humans perceiving visual scenes as objects, Object-Centric Learning (OCL) can abstract dense images or videos into sparse object-level features. Transformer-based OCL ha…