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
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.
FLEET is a token‑based feature extractor that processes event camera data directly, using random Fourier features and cross‑attention to compress variable‑length event streams into fixed‑size latent representations. By decoupling inference cost from sensor resolution, it avoids the high compute and temporal blurring associated with CNN‑based grid aggregation. Experiments on a new high‑throughput benchmark show that FLEET outperforms state‑of‑the‑art methods and remains robust across different observation frequencies.
By Tristan Gottwald, Maximilian Schier, Melanie Schaller, Bodo Rosenhahn
The paper introduces Gated Token Recurrence (GTR), a softmax‑free recurrent vision backbone that replaces global softmax attention with gated linear attention, alternating scan directions, and enhanced SwiGLU blocks. GTR is distilled from a DINOv3 teacher using only final‑layer patch‑token alignment, and achieves strong performance on COCO object detection (58.9 box AP) with very low latency (1.908 ms on an RTX 4090). The backbone also transfers to multiple dense prediction tasks and runs efficiently on edge hardware via a specialized CUDA operator and TensorRT deployment.
By Zhe Feng, Longfei Liu, Wei Liu, Kai Chen, Jiangjiang Kong, Wei Zhou, Yifeng Qian, Dexiong Chen, Xuanlong Yu, Xi Shen
The paper introduces a confidence‑normalized continuous multi‑timescale representation for event cameras, using logarithmic B‑spline temporal encoding and a geometry‑aware local confidence mechanism. When paired with a fixed feed‑forward EventCenterNet detector, this representation outperforms the compact CSTR representation on the PEDRo and Gen1 datasets. Additionally, a recursive exponential‑polynomial approximation is proposed to allow efficient event‑by‑event updates while maintaining detection performance.
By Fredrik Lundell, Per-Erik Forssen, M{\aa}rten Wadenb\"ack, Astrid Lundmark
PointEvent introduces serialized motion evidence accumulation for event-based tiny object detection, treating motion continuity as an ordered evidence propagation process. The method organizes event streams into locality‑preserving spatiotemporal paths and chronology‑preserving temporal paths, alternating serialized scans across complementary orders to consolidate fragmented motion evidence. A lightweight event‑wise state‑space framework with a high‑resolution event branch and compact context modulation achieves state‑of‑the‑art performance with the fewest parameters and fastest inference among compared methods.
By Zongze Wu, Baofeng Jia, Weiqi Yan, Jingyuan Zhang, Yu Zang, Xiaoyu Chen, Jing Han
Event cameras generate asynchronous, high-frequency data streams offering spatially sparse information at lower latency than traditional cameras. In principle, these properties should be ideal for the design of control policies.
BlazeVGGT is a training‑free, plug‑and‑play module that speeds up the VGGT visual geometry model by sparsifying cross‑view attention. It selects a compact set of complementary views using a greedy determinantal point process, preserving dense per‑frame predictions while eliminating redundant computations. On 2,000‑frame sequences, BlazeVGGT achieves roughly 13× faster inference and improves Chamfer Distance from 0.41 to 0.32, demonstrating a strong accuracy–efficiency trade‑off for point‑cloud reconstruction and camera‑pose estimation.
By Shujun Yao, Junhui Lin, Congying Han, Tiande Guo, Ziwen Liu, Bonan Li
arXiv:2609.38984v1 Announce Type: cross
Abstract: World-action models (WAMs) leverage pretrained video models to improve generalization in robot control by jointly predicting future visual states and...
By Xinling Xie, Haodong Wang, Jiazhi Mi, Zhiming Liu, Zicong Hong, Xiaoyi Pang, Qianli Liu, Yangjia Hu, Ying Chen, Zhengyang Yan, Song Guo
PPS: Plug-and-Play Saccadic Vision for Fine-Grained Classification proposes a lightweight extension that transforms any pretrained backbone with spatial feature maps into a saccadic classifier. The method adds only a small priority head and a single cross-attention fusion block, using a weight-tied encoder to process a downsampled peripheral glance and high-resolution glimpse crops selected by a priority distribution. A sequential sampler with non-maximum suppression draws a configurable number of glimpses, and a residual multi-head cross-attention block fuses the glance with the glimpses; the priority head is trained via a per-glimpse information-gain target from a lightweight memory bank, avoiding reinforcement learning. The approach achieves linear complexity in the number of glimpses and decouples peak memory from input resolution, consistently improving performance across nine fine-grained classification benchmarks.
By Johann Schmidt, Sebastian Stober, Joachim Denzler, Paul Bodesheim
arXiv:2605.15195v2 Announce Type: replace
Abstract: Recent feed-forward reconstruction models, such as VGGT, have proven competitive with traditional optimization-based reconstructors while also prov...
By Jianyuan Wang, Minghao Chen, Shangzhan Zhang, Nikita Karaev, Johannes Sch\"onberger, Patrick Labatut, Piotr Bojanowski, David Novotny, Andrea Vedaldi, Christian Rupprecht
arXiv:2609.17134v1 Announce Type: new
Abstract: Neuromorphic vision systems operate under strict constraints on bandwidth, memory, and energy, particularly at the edge, motivating early mechanisms fo...
By Luca Peres, Giulia D'Angelo, Chiara Bartolozzi, Oliver Rhodes