Semantic Slots for Video Object-Centric Learning proposes a Transformer-based decoder that frees slots from spatial anchoring, enabling them to act as semantic queries invariant to object position. This design removes the need for complex temporal predictors, allowing a single-frame slot to decompose subsequent video frames. On YouTube-VIS, the method outperforms prior approaches, improving mBO by 31 points over VideoSAUR and achieving state-of-the-art results with 86.6% ARI and 62.8% mBO.
By Khalil Sabri, Guillaume-Alexandre Bilodeau, Nicolas Saunier, Wassim Bouachir
The paper introduces Skeleton-Language feature Pooling Switching, a weakly‑supervised vision‑language pretraining strategy for skeleton‑based zero‑shot spatio‑temporal action localization. It replaces video‑level pooling with instance‑level feature computation during inference, enabling the model to estimate unseen actions without costly annotations. Additionally, Scene‑Mixed Discriminative Contrastive Learning is proposed to separate actions at the instance level within mixed scenes using a MIL framework, and experiments on four public datasets confirm the method’s effectiveness.
By Koshiro Nagano, Fumiaki Sato, Ryo Hachiuma, Kazuki Tsutsukawa, Taiki Sekii
arXiv:2608.30388v1 Announce Type: cross
Abstract: Cross-view video representation learning aims to capture viewpoint-invariant action semantics despite substantial appearance changes across egocentri...
By Youngchae Chee, Hosu Lee, Sungjune Park, Junho Kim, Yong Man Ro
Fine-grained understanding of operating room (OR) activity could enable workflow-aware assistance, yet remains difficult due to clutter, occlusions, and limited sensing. The prevailing approach to model this environment is scene graphs as an interpretable representation of OR interactions.
SV‑TAD introduces native sparse 2D convolutions to enable convolutional adapters to operate directly on dynamically pruned token sets, addressing the inefficiency of existing adapters that do not reduce inference‑time computation. By integrating this primitive, SV‑TAD cuts VideoMAEv2‑L computation by up to 64% and achieves 2.2× faster inference while preserving state‑of‑the‑art accuracy on THUMOS‑14 and ActivityNet‑1.3. When applied to InternVideoNext‑L, the approach surpasses prior state‑of‑the‑art performance at roughly half the computational cost and also supports auxiliary task tokens, improving fine‑grained assembly detection on ATTACH.
arXiv:2505. 20894v2 Announce Type: replace Abstract: Despite recognized limitations in modeling long-range temporal dependencies, Human Activity Recognition (HAR) has traditionally relied on a sliding window approach to segment labeled datasets.
By Marius Bock, Juergen Gall, Michael Moeller, Kristof Van Laerhoven
SV-TAD introduces native sparse 2D convolutions to enable convolutional adapters to operate directly on dynamically pruned token sets, addressing the inefficiency of dense reconstruction in token‑selection methods. By integrating this primitive into a temporal action detection framework, the approach cuts VideoMAEv2‑L computation by up to 64% and achieves 2.2× faster inference while preserving state‑of‑the‑art accuracy on THUMOS‑14 and ActivityNet‑1.3. When applied to InternVideoNext‑L, it surpasses prior state‑of‑the‑art performance at roughly half the computational cost and naturally supports auxiliary task tokens for improved fine‑grained assembly detection on ATTACH.
By Ricardo Pizarro, Roberto Valle, Jos\'e M. Buenaposada, Luis M. Bergasa, Luis Baumela
arXiv:2608.27562v1 Announce Type: new
Abstract: Translating continuous, noisy egocentric video streams into discrete, temporally ordered action steps is fraught with visual challenges. Heavy ego-moti...
By Anubhav Gupta, Archit Kambhamettu, Vatsal Agarwal, Pulkit Kumar, Abhinav Shrivastava
The paper introduces Latent-Action-Guided Video-Language Feature Learning (LAG-VLFL) for recognizing surgical instrument–tissue interactions. By compressing frame-to-frame feature changes into latent actions and predicting next‑frame features, the method aligns video and textual action descriptions without requiring extra spatial or motion annotations. Experiments show that LAG-VLFL improves interaction grounding, temporal‑direction sensitivity, and achieves competitive recognition with faster inference and lower INT4 accuracy loss compared to V‑JEPA2/2.1.
By Jiajun Cheng, Sainan Liu, Subarna Tripathi, Xiaofan Yu, Shan Lin
arXiv:2401. 10805v4 Announce Type: replace-cross Abstract: We introduce the novel concept of visually Connecting Actions and Their Effects (CATE) in video understanding.
By Paritosh Parmar, Eric Peh, Basura Fernando
The paper presents a new multi-frame representation for event-based action recognition that outperforms single-frame approaches, especially on long and complex event sequences. It employs a masked auto-encoding framework to pre-train a Vision Transformer, using a sparsity-aware masking algorithm, token selection, and multiple mask tokens with position-aware attention to boost performance while managing computational cost. The method achieves state‑of‑the‑art results on the THU‑EACT‑50 CHL and DailyDVS‑200 benchmarks.
By Emma Norrby, Tomas Wilkinson, Per Nyblom, David Gustafsson
Precise Event Spotting (PES) requires distinguishing visually similar yet semantically distinct adjacent frames, making it fundamentally different from image classification and coarse action recognition. Although self-distillation methods such as DINO have shown strong representation learning ability in images, we find that directly applying them to PES is ineffective: without supervised guidance, subtle but crucial motion cues are often suppressed as noise, leading to representations that are insensitive to precise event boundaries.