Detection, segmentation, depth and recognition research, plus the vision backbones that keep displacing the last generation.
arXiv:2609.08318v1 Announce Type: cross Abstract: The transition from human-centric assistance to Autonomous Software Engineering (ASE) agents has enabled the resolution of complex real-world SE task...
The paper introduces MedSegLatDiff, a diffusion-based framework that combines a variational autoencoder (VAE) with a latent diffusion model for medical image segmentation. By compressing images into a low-dimensional latent space, the method reduces noise and speeds up training, while a weighted cross‑entropy loss preserves tiny structures such as small nodules. Evaluated on ISIC‑2018, CVC‑Clinic, and LIDC‑IDRI datasets, MedSegLatDiff achieves state‑of‑the‑art Dice and IoU scores, generates diverse segmentation hypotheses, and produces confidence maps that enhance interpretability and reliability for clinical deployment.
arXiv:2609.06165v1 Announce Type: cross Abstract: Whole-body diffusion-weighted imaging (WB-DWI) is widely used for multiple myeloma (MM) assessment, yet automated lesion segmentation remains challen...
arXiv:2609.06581v1 Announce Type: cross Abstract: Small-object detection remains challenging because limited pixels cause information loss and suppress the scale knowledge encoded in pretrained detec...
arXiv:2510.10471v3 Announce Type: replace-cross Abstract: Environmental perception systems are crucial for high-precision mapping and autonomous navigation, with LiDAR serving as a core sensor provid...
BlueprintAgent (BPA) is a multimodal agent that converts scanned reinforced‑concrete building blueprints into simulation‑ready frame models by treating a large language model (MLLM) as the primary reader and decision maker, while OCR and computer vision provide localized evidence. BPA implements engineering constraints as callable validators that trigger targeted MLLM revisits over specific regions, enabling precise beam‑column support, span count, and 3D continuity checks. In evaluations on 300 real scanned sheets from 20 projects, BPA achieved a macro‑averaged Beam F1 of 0.994, vastly outperforming single‑MLLM zero‑shot (0.301) and fixed‑pipeline (0.820) baselines.
The study introduces the RSM framework, which includes the ARNAI artifact removal network, to enhance automated measurement of spinopelvic parameters on postoperative lumbar spine radiographs containing implants. Adding ARNAI to the Transformer-based FCBFormer model raised the mean Dice similarity coefficient from 0.814 to 0.870 and significantly reduced the mean L4–L5 segmental Cobb angle error from about 15.8° to 4.7°, a 70% improvement. The framework also improved intraclass correlation coefficients for key parameters, surpassing 0.70 for pelvic tilt, lumbar lordosis, and sacral slope.
AGSA-Net is a hyperspectral image classification framework that incorporates spectral unmixing priors through an abundance-guided self‑attention network. It first estimates physically meaningful subpixel abundance maps with non‑negativity and sum‑to‑one constraints, then uses these abundances to build an affinity prior that directs a spectral transformer to focus on class‑discriminative interactions. The transformer features are fused with compact abundance descriptors for final classification, and experiments on Indian Pines, Augsburg, and Berlin datasets show improved performance, especially in heterogeneous urban scenes.
The study investigates whether Vision Transformer (ViT)-based animal re-identification models learn biologically meaningful concepts. Using a DINOv3 backbone fine‑tuned on Western lowland gorilla images, the authors find that sex and age emerge as linear directions in the model’s representations, generalizing to unseen individuals with high AUROC scores. They demonstrate that the sex direction is causally used by the model, that fine‑tuning relocates these concepts within the network, and that the representations reflect a graded biological axis encoded redundantly across the population.
AtomCite is an agentic framework that verifies and corrects page‑level citations in multi‑page documents by parsing answers into claims, checking each claim against the cited page image, and applying a deterministic repair policy. The authors introduce DocCite, the first benchmark for this task, built on MP‑DocVQA and DUDE, containing 928 injected instances and 1,909 verified natural errors. Across Gemini, Claude, and GPT models, AtomCite achieves about 93% verification accuracy and improves citation precision from 34% to 87‑90%, while also enhancing hallucination detection in open‑source models.
The paper studies how Bird's‑Eye‑View (BEV) maps predicted by Cross‑View Transformers (CVT) can be used directly as inputs to a Behavior‑Cloning (BC) driving policy in the CARLA simulator. It introduces a six‑channel BEV representation and a Kernel Density Estimation (KDE) weighting scheme to focus learning on underrepresented maneuvers. Closed‑loop tests show that the KDE‑weighted model is the only predicted‑BEV agent to finish an episode without infractions, highlighting that global segmentation scores are poor proxies for driving performance and that prediction quality at critical geometries, especially the route channel, is key to reliable navigation.
The paper presents a family of tunable lifting schemes for biorthogonal wavelet filter banks, introducing three strategies: low‑pass tuning, high‑pass tuning, and a sequential scheme that jointly adapts both branches. These designs use a lattice‑based lifting structure to ensure invertibility and stability for any parameter values. The authors integrated the schemes into a ResNet‑18 backbone and evaluated them on image classification (DTD) and anomaly detection (MVTec‑AD and KRC102S), reporting consistent performance improvements across all tasks.
The paper presents a solution for the UCF UrbanTwin LUMPI Track in the Sim-to-Real LiDAR Challenge, where a detector trained solely on synthetic data must perform on real LiDAR frames. The approach tackles the Sim2Real gap through data alignment, diversified sampling, augmentation, and specialized detectors, followed by class-aware fusion and calibration techniques. The final submission achieved a Combined Score of 0.4692, a Detection Score of 0.1797, a Realism Score of 0.9035, and a 3D mAP@0.5 of 0.1258.
The paper investigates how the choice of segmentation affects LIME explanations for document image classification. It compares standard superpixel methods (Quickshift and SLIC) with document-aware segmentations based on OCR bounding boxes and regular grids on the RVL-CDIP dataset. The study finds that document-aware segmentations yield more consistent, correct, and locally faithful explanations, require fewer perturbations, and reveal shortcut behaviors tied to document identification codes that superpixel-based LIME often hides.
arXiv:2605.13214v3 Announce Type: replace-cross Abstract: Recent cryptographic results establish that neural networks can be backdoored such that no efficient algorithm can distinguish them from a cl...
The paper introduces a structure‑aware federated learning framework for segmenting catheters and guidewires in X‑ray fluoroscopy. It presents a new benchmark dataset, CathAction, and a shape‑sensitive loss that improves segmentation accuracy. The approach extends to federated learning, adding projected gradient descent for adversarial optimization, and includes a diffusion‑based synthetic data generator that boosts performance under data scarcity.
Vague2Detect is a hybrid pipeline that improves object detection for ambiguous prompts by combining a fine‑tuned Sentence‑BERT to retrieve candidates from a structured household knowledge base, YOLO‑World to verify their presence in images, and a GPT‑3.5‑turbo fallback to generate new candidate descriptions when prompts fall outside the knowledge base. On a benchmark of household scenes, Vague2Detect raises the vague prompt success rate from 32% (YOLO‑World alone) to 61% with high precision, and up to 85% when the GPT fallback is used.
The paper describes a system for the SHROOM-Visions 2026 shared task on character-level VLM hallucination detection. It combines a small 4‑B‑parameter VLM fine‑tuned as a per‑token classifier that uses a two‑token feature from its hidden states with a large ~400‑B zero‑shot VLM judge at prediction time, both leveraging OCR of visible in‑image text. Using synthetic hallucination data from the large model for ensemble diversity and validation‑based selection of feature layer, training data, and OCR grounding, the entry achieved competitive results across multiple languages.
AudioFuse is a hybrid architecture that jointly learns from spectrograms and raw waveforms to classify phonocardiograms. It combines a wide-and-shallow Vision Transformer for spectral features with a shallow 1D CNN for temporal waveforms, reducing overfitting while capturing complementary information. On the PhysioNet 2016 dataset, AudioFuse achieves a state‑of‑the‑art ROC‑AUC of 0.8608 and shows superior robustness to domain shift on the PASCAL dataset, outperforming both spectrogram‑only and waveform‑only baselines.
GoDeep is an annotation‑free method for open‑vocabulary 3D scene understanding that uses a vision‑language model solely as a translator to generate structured, entity‑level descriptions of each image. These descriptions are projected and aggregated in a language‑only embedding space, eliminating the need for a 3D training corpus or domain‑specific encoder. The approach achieves competitive performance on ScanNet++ and a cultural heritage benchmark, accurately localizes out‑of‑vocabulary objects, and offers explainable, point‑level predictions.