SFVO is a stereo visual‑odometry framework that leverages pretrained stereo‑matching and optical‑flow models to obtain dense stereo and temporal correspondences. Rather than learning pose directly from images, it maps these correspondences into geometric constraints and predicts trustworthy points using decoupled confidence maps for rotation and translation. Experiments on both outdoor and indoor datasets show that SFVO delivers robust, accurate pose estimation with strong generalization, and the authors plan to release the code.
By Kai Zhang, Guoyang Zhao, Jun Ma
EventGeM introduces a global‑to‑local feature fusion pipeline for event‑based visual place recognition, combining whole‑image feature detection with 2D homography‑based re‑ranking via RANSAC. It adds a regional generalized mean (GeM) pooling layer that learns to extract the most relevant spatial features from event streams, producing a compact global descriptor trained on the NYC‑Event‑VPR dataset. The method demonstrates significant improvements in viewpoint‑robust localization, achieving 7–43 percentage point gains in Recall@1 over the strongest baseline and real‑time performance on a robotic platform.
By Adam D. Hines, Gokul B. Nair, Nicol\'as Marticorena, Michael Milford, Tobias Fischer
The paper demonstrates that object detection benchmarks suffer from incomplete annotations, with re-annotation of COCO, Pascal VOC, Cityscapes, and KITTI revealing up to a 60% increase in detected objects, especially small, occluded, or densely packed instances. The authors propose a scalable annotation pipeline that uses multiple annotators per object to capture uncertainty and improve recall, and they introduce two new large-scale benchmarks: an uncertainty-aware detection benchmark and a label error detection benchmark based on real errors. Their findings show that benchmark performance is highly sensitive to annotation quality, yet model rankings remain largely unchanged, highlighting the need for uncertainty-aware evaluation to better reflect real-world ambiguity.
By Sarina Penquitt, Jonathan Klees, Antonia van Betteray, Parssa Jashnieh, Peter Stehr, Matthias Rottmann, Lars Schmarje
The paper introduces Geometric Mean Pooling (GMP), a signed pooling operator that combines the product of feature signs with the geometric mean of feature magnitudes. GMP preserves joint sign information and a multiplicative scale without learnable parameters, and it is shown to outperform average and max pooling on synthetic sequence tasks and maintain performance under multiplicative noise. However, its effectiveness on image and molecular data depends on representation, target parameterization, and pooling placement, positioning GMP as a complementary, regime‑dependent inductive bias rather than a universal replacement for standard pooling.
By Ang-Kun Wu, Fangdi Wen, Jingtao Zhang
VoxelTTO is a feed‑forward framework that reconstructs 3D Gaussian splatting scenes from multiple images by aggregating dense image features into a global voxel representation and decoding Gaussians from voxel features, thereby eliminating the pixel‑to‑Gaussian correspondence. It incorporates test‑time optimization with lightweight LoRA modules to adapt to known camera parameters while keeping the pretrained visual foundation model frozen. The method replaces standard rasterization with stochastic solid volume rendering, improving geometric fidelity, and demonstrates superior RGB‑D novel‑view synthesis and camera‑pose estimation on Replica, Tanks and Temples, and DTU datasets.
By Yibin Zhao, Yihan Pan, Yangwen Li, Jun Nan, Jianjun Yi
LoG-VGGT is a memory‑efficient framework for long‑sequence 3D reconstruction that balances local temporal modeling with global camera consistency. It uses cross‑window attention in a small subset of transformer blocks to propagate information across adjacent temporal windows while keeping memory usage bounded. A global camera consistency refinement module further improves long‑horizon pose stability by enforcing scene‑level constraints through cross‑attention between camera and compact register tokens, leading to better depth accuracy and robust camera pose estimation on multiple benchmarks.
By Jingke Zhou, Chenhang Ma, Zhizhou Zhong, Mingkai Liu, Zhuang Zhou, Yicheng ji, Binghua Su, Bo Cai, Xianliang Huang
arXiv:2609.21938v1 Announce Type: new
Abstract: Transformer-based models have recently achieved strong performance on 3D reconstruction from images, and recent works extend them to process video stre...
By Sunghyun Baek, Hanna Bae, Minchan Kwon, Junmo Kim
The paper introduces a Chinese competitive debating dataset and benchmark, comprising 148 fully adjudicated matches with 2,698 stages and 20,542 exchange units, all annotated by 120 professional judges using a shared rubric. It offers manually verified transcripts, segmentation, and preserved scores, votes, and rationales, enabling evaluation of large language models on winner-tendency prediction, stage-score prediction, and best-debater prediction. Zero‑shot experiments show the best models achieve 66.2% winner‑prediction accuracy, a 0.250 Pearson correlation for stage scores, and 56.8% best‑debater accuracy.
By Zongrui Yang, Haoyuan Li, Zhongsheng Wang, Zhirui Zeng, Pengqian Han, Yi Zhou, Yuting Wang, Jiamou Liu
arXiv:2609.21903v1 Announce Type: new
Abstract: Leaf-wood segmentation of individual trees from LiDAR point clouds is essential for quantitative structure models (QSMs) used in non-destructive biomas...
By Roman Kaharlytskyi, Derek T. Robinson, Roberto Guglielmi
arXiv:2609.21698v1 Announce Type: new
Abstract: Deep learning segmentation of renal tumors requires large annotated datasets, yet clinical deployments typically offer only a handful of tumor-positive...
By Ekaterina Sedykh, Salme Ussanov, Dmytro Fedorenko, Dmytro Fishman
arXiv:2609.21597v1 Announce Type: new
Abstract: Monocular 6-DoF pose estimation of non-cooperative targets is important for on-orbit servicing and debris removal. A single-image estimator can confuse...
By Andr\'e Lopo, Atabak Dehban, Rodrigo Ventura
The paper compares Ultralytics YOLO27, YOLO26, YOLO11, and YOLOv8 for detecting and segmenting small fruit parts in orchard settings. It evaluates five model scales across 30 experiments, finding that YOLO11s-960 and YOLO26s-960 achieve the best mask and box mAP scores while maintaining efficient parameter counts. The study also highlights the difficulty of peduncle detection and provides publicly available code and models for reproducibility.
By Ranjan Sapkota, Manoj Karkee
IntBMoE introduces a block‑conditioned mixture‑of‑experts that decouples participation, execution, and materialization by combining dense expert composition with sparse block execution. Each internal layer uses a lightweight hypernetwork to merge all expert bases into a single composed expert, while a router selects only a few blocks per token, keeping compute and memory costs low. Experiments on image classification, language modeling, and sequential recommendation demonstrate consistent performance gains, and the model is deployed in AMap’s generative recommendation system, improving UVCTR by 2.4% in online A/B tests.
By Ran Cheng, Longfei Xu, Zheng Liu, Kaikui Liu, Xiangxiang Chu
Monocular colonoscopic 3D reconstruction is important for surgical robotic colonoscopy, but remains challenging due to weak texture, specular reflections, limited view overlap, and non-rigid tissue mo...
My comment on MCP was always a bad idea? — Hacker News.This article entirely misses the value that MCP brings today.
Sure, there's almost no reason to use MCPs if you are running a full-blown te...
We present M3GA-Wild, the first benchmark for multi-modal, multi-session ground-to-aerial place recognition in forests. M3GA-Wild unifies and extends existing forest localisation datasets, providing a...
The paper introduces Riemannian–Lorentz Parameter Fusion (RLPF), a method for merging a Vision Transformer and a state‑space model without gradient descent. RLPF aligns parameter groups by semantic role, projects them onto a common coordinate system, lifts selected coordinates to the Lorentz hyperboloid, computes a regularized geodesic barycenter, and decodes the result back into the two branches, with a learned gate combining their logits. The resulting fine‑tuned system achieves 82.37 % on CIFAR‑10, 75.04 % on Oxford‑IIIT Pet, and 78.58 % top‑1 accuracy on ImageNet‑1K, surpassing the best‑parent accuracies of 76.54 %, 71.42 %, and 76.42 % respectively.
By Badri N. Patro, Vijay S. Agneeswaran
This study evaluates the classification accuracy of six modern deep‑learning architectures—VGG19, ResNet50, GoogleNet, ConvNeXt, EfficientNet, and Vision Transformers—on breast ultrasound images categorized by BI‑RADS. Using 2,945 training images and 936 validation images from 1,540 patients, the models were tested in full fine‑tuning, linear evaluation, and training‑from‑scratch settings. The best performance was achieved with full fine‑tuning, yielding 76.39 % accuracy and a 67.94 % F1 score.
By Malitha Gunawardhana, Norbert Zolek
The paper introduces LAMP, a foreground‑guided framework for synthesizing realistic polyp images that preserves the supplied lesion while generating compatible mucosal tissue. LAMP separates the lesion, valid mucosa, and camera exterior using a field‑of‑view mask, employs lesion‑to‑mucosa cross‑attention to extract appearance conditions, and propagates these conditions over legal tissue support via a FOV‑constrained multidirectional Vision Receptance Weighted Key‑Value mechanism. Experiments on five polyp datasets show that LAMP outperforms existing methods in generation quality and consistently improves five downstream segmentation models.
By Tong Wang, Yuting He, Bin Ren, Yutong Xie, Guanyu Yang
The Public Discourse Corpus (PDC) is the first dataset of public‑figure interview speech annotated for affective valence and epistemic modality. It contains 998 videos from 100 speakers across seven professional domains, yielding 186,642 sentences (3.1 million words). A key methodological contribution is Target Speaker Participation (TSP), a five‑category annotation taxonomy with documented inter‑annotator reliability (κ = 0.616), and an audio‑first diarization pipeline that separates target‑speaker turns from interviewer and third‑party speech. The corpus, annotation tools, validation sample, and processing pipeline are released as open source.
By Bo Chen