Detection, segmentation, depth and recognition research, plus the vision backbones that keep displacing the last generation.
arXiv:2609.08330v1 Announce Type: new Abstract: Table detection is a core task in document analysis, supporting downstream applications such as information retrieval, document reconstruction, and vis...
arXiv:2609.07072v1 Announce Type: cross Abstract: Graphic designs, such as posters, advertisements, and infographics, are an important medium for communicating information and shaping understanding....
The Living Library is an end‑to‑end framework that converts fragmented digital archives into governed, conversational exhibit experiences. Developed at the Theodore Roosevelt Presidential Library, it digitizes a 300,000‑record collection, enriches it with OCR and metadata, and publishes it to a hybrid dense/semantic index. The system supports curator review via the Archivist App, powers a researcher interface, and runs Talk to TR—a museum exhibit where a digital human embodiment of Theodore Roosevelt answers visitors’ questions using Cross‑Era Analogical Grounding and dual‑path retrieval to keep responses grounded and responsive.
The paper introduces TITAnD, a Trajectory Image Transformer that converts dense and sparse GPS trajectories into a Hyperspectral Trajectory Image (HTI) and applies vision-based classification and segmentation for anomaly detection. It employs a Cyclic Factorized Transformer (CFT) that splits attention along within-day and across-day axes, drastically reducing computational cost and enabling multi-month analysis. Empirical results show TITAnD outperforms existing sparse and dense benchmarks, achieving higher AUC-PR and faster inference than comparable Transformers.
arXiv:2605.00809v3 Announce Type: replace Abstract: In this paper, we present \textbf{Gen}erative \textbf{L}anguage-\textbf{I}mage \textbf{P}re-training (GenLIP), a minimalist generative pretraining...
arXiv:2609.05662v1 Announce Type: cross Abstract: Full-page end-to-end Optical Music Recognition seeks to transcribe entire music pages directly into symbolic notation, avoiding the limitations of tr...
arXiv:2609.05516v1 Announce Type: cross Abstract: Unified perception enables autonomous driving systems to perform object detection, drivable-area segmentation, and lane segmentation within a single...
arXiv:2605.21957v2 Announce Type: replace Abstract: Video anomaly detection is critical for public safety and security, yet remains highly challenging despite extensive research due to large variatio...
arXiv:2609.07313v1 Announce Type: new Abstract: Segmentation of complex structures in X-ray tomographic data is a fundamental task in biomedical research, but it often requires large amounts of preci...
High-resolution water mapping supports environmental monitoring and related applications, but accurate pixel-level labels are difficult and costly to produce. Official hydrographic vectors provide sca...
Small object detection in unmanned aerial vehicle (UAV) and remote sensing imagery requires preserving high-resolution detail while modeling long-range context. Adding a stride-4 detection level and r...
Real-world detectors must often interpret functional or ambiguous prompts, yet conventional models such as YOLO remain restricted to fixed class lists. Even open-vocabulary models like YOLO-World freq...
SeGDeP introduces a decoupled prompt system that separates semantic identification from spatial grounding in reasoning segmentation. It uses a semantic prompt branch and an independent geometric projection path to produce semantic features and a DETR‑predicted box, which jointly condition a SAM mask decoder. The model achieves high cIoU and gIoU scores while adapting only a small fraction of Qwen3‑VL parameters via LoRA.
TriCCOT is a tri-part architecture designed for onboard space object detection that balances computational efficiency with robust performance. It combines a convolutional region proposal network, a conformal prediction stage that enlarges bounding boxes with distribution‑free probabilistic coverage, and Aper‑GATES, a hardware‑friendly attention‑based classifier that replaces standard transformer operations with convolutional projections and gating. Experiments on DIOR and VDVRaw datasets show competitive detection accuracy and improved robustness to spatial blur and noise, and the model was fully deployed on a Xilinx Versal VCK190 FPGA without altering the underlying DPU architecture.
MARS-CLIP is a zero‑shot semantic segmentation framework that builds on CLIP by adding a multi‑resolution feature extraction module and an attention refinement mechanism. The multi‑resolution module fuses fine‑grained local features with global context to mitigate low spatial resolution, while the attention refinement injects spatial and color biases from intermediate layers into the final self‑attention block to better recover object boundaries. Experiments on six public datasets show that MARS‑CLIP outperforms state‑of‑the‑art methods.
We present our solution to the LUMPI track of the UCF UrbanTwin Sim2Real LiDAR Challenge at the 6th DriveX Workshop, ECCV 2026. The detector must be trained only on synthetic data and is evaluated on...
The paper introduces a semi‑supervised hyperspectral image classification framework that combines spatial prior information with a dynamic learning mechanism. It proposes an Edge‑Aware Superpixel Label Propagation module to reduce boundary label diffusion and a Dynamic History‑Fused Prediction method to stabilize pseudo‑labels over time. Additionally, an Adaptive Tripartite Sample Categorization strategy is used to hierarchically exploit easy, ambiguous, and hard samples, resulting in improved pseudo‑label quality and learning efficiency. The combined Dynamic Reliability‑Enhanced Pseudo‑Label Framework achieves spatio‑temporal consistency optimization and demonstrates superior performance on four benchmark datasets.
The paper introduces a biologically inspired framework that learns object‑centric visual representations from raw videos without human annotations or camera calibration. By using motion boundaries detected via optical flow and clustering to create pseudo‑instance masks, the method supervises a single‑image encoder with pixel‑level pairwise metric learning. Training on 195 million pseudo‑labeled frames and expanding to 421 million frames through Motion‑Verified Self‑Training, the approach yields Swin‑based encoders that outperform or match supervised and self‑supervised baselines on tasks such as monocular depth estimation, 3D object detection, 3D occupancy prediction, and end‑to‑end planning.
This study presents a new mobile‑sensing dataset from Surat, India, capturing PM2.5 concentrations along with meteorological and land‑use variables. The authors model the data as a graph using two node‑definition strategies—uniform segmentation and DBSCAN clustering—and introduce a Spatially Attentive Graph Neural Network (SA‑GNN) that combines cluster‑specific GRUs with a Graph Attention Network to forecast fine‑grained, short‑term PM2.5 levels. SA‑GNN outperforms traditional LSTM, RNN, GRU, and ANN baselines, achieving an R² of 0.95, RMSE of 6.8, and MAE of 4.2 µg/m³ on the dataset.
AnchorWeave is a memory‑augmented video generation framework that replaces a single misaligned global memory with multiple clean local geometric memories. It retrieves local memories based on coverage of the target trajectory and integrates them through a multi‑anchor weaving controller during generation. Experiments show that this approach improves long‑term scene consistency while preserving visual quality, with ablation studies confirming the benefits of local conditioning, multi‑anchor control, and coverage‑driven retrieval.