CapsuleMotion is a lightweight, real‑time visual motion predictor designed for video capsule endoscopy (VCE). It predicts motion between successive frames using on‑device image compression metrics, allowing the capsule to adjust its frame rate dynamically and operate in a low‑power mode before entering the small intestine. Evaluated on the Rhode Island VCE dataset and deployed on an ultra‑low‑power RISC‑V demonstrator, CapsuleMotion reduces energy consumption by up to 20.66% and improves detection of the small intestine entry point.
arXiv:2310. 07895v2 Announce Type: replace Abstract: This paper presents a method to efficiently classify the gastroenterologic section of images derived from Video Capsule Endoscopy (VCE) studies by exploring the combination of a Convolutional Neural Network (CNN) for classification with the time-series analysis properties of a Hidden Markov Model (HMM).
UnCapsTSR is an unsupervised transformer-based GAN framework designed to enhance the spatial resolution of low‑resolution wireless capsule endoscopy (WCE) images. It eliminates the need for explicit degradation modeling or paired LR‑HR data by using a Bilateral Total Variation loss to preserve spatial continuity. The authors introduce a new Kvasir Capsule dataset for training, validate generalizability on KID and GIANA datasets, and propose the Endoscopy Quality Metric (EndoQM) as a non‑reference evaluation tool, reporting 40–80% improvement in EndoQM over state‑of‑the‑art unsupervised methods.
The study evaluates a 2.5D U‑Net model for detecting gaseous microemboli (GME) in real‑time during cardiac surgery using transesophageal echocardiography (TEE). On a pilot dataset of eight patients, the model achieved high precision (92.55%) and recall (80.54%) with an average inference time of 0.12 s per batch, outperforming classical spot detection and 2D U‑Net while maintaining real‑time speed. External validation on a GME‑negative dataset showed few false positives, supporting the model’s feasibility for real‑time GME segmentation.
arXiv:2608. 03430v1 Announce Type: cross Abstract: Four-dimensional cone beam CT (4D CBCT) is important for image-guided radiation therapy of thoracic cancers, but its use is limited by long scan times, causing high patient dose and motion/sparse-sampling artifacts.
CEM‑TUDASR is a lightweight, unsupervised Transformer-based super‑resolution framework designed to enhance low‑resolution images from Wireless Capsule Endoscopy (WCE). It uses a domain‑adaptive degradation network to generate realistic WCE‑like low‑resolution images from high‑resolution conventional endoscopy data, enabling effective unpaired learning. The model incorporates Deep Attention Blocks and a Fusion Attention Block to capture both global context and fine local details, achieving superior performance on WCE datasets and demonstrating cross‑domain adaptability to retinal images, all while keeping the parameter count and computational load low.
CEM‑TUDASR is a lightweight, unsupervised Transformer‑based super‑resolution framework designed for Wireless Capsule Endoscopy (WCE) images. It uses a domain‑adaptive degradation network to synthesize realistic low‑resolution WCE images from high‑resolution conventional endoscopy data, enabling unpaired training. The SR generator incorporates Deep Attention Blocks and a Fusion Attention Block to preserve both global context and fine local structures, achieving superior no‑reference quality metrics and improved restoration of mucosal textures, vascular patterns, and anatomical details while remaining computationally efficient.
arXiv:2608. 13711v1 Announce Type: cross Abstract: Computer-aided detection (CADe) systems for colonoscopy promise to reduce clinical miss rates, yet reliable real-world deployment remains elusive.
The paper introduces a lightweight full-frame detector for partially manipulated AI-generated videos, suitable for edge deployment without face-detection preprocessing. It distills a DINOv2-Base teacher into a frozen MobileNetV3-Small student using temperature-annealed soft-label transfer, attention-diversity regularization, frame-level supervision, and a residual feature adapter. The model addresses false positives on legitimate scene cuts and threshold-level miscalibration, achieving an AUC of 0.766 on a 55,393-sample spliced test set while running at 3.65 ms per 16‑frame clip with a 150.4 MB checkpoint.
Albireo is an adaptive, energy‑efficient inference framework for video object detection on edge devices that wraps existing detectors without modification. It uses a 10‑dimensional Kalman filter per active object to decide when to skip detector calls, predicting bounding boxes on skipped frames at near‑zero GPU cost. Evaluated on BDD100K with YOLO and RF‑DETR detectors on NVIDIA Jetson AGX Thor and Orin, Albireo maintains AP@50 within ±1.2 pp of full‑frame inference while reducing energy consumption by 12.1–17.6 % and improving accuracy for some models.
arXiv:2602. 06938v2 Announce Type: replace-cross Abstract: The classification performance of deep neural networks relies strongly on access to large, accurately annotated datasets.
RAUL is a reference‑assisted reconstruction framework that recovers ureteroscope trajectories from endoscopic video alone, using a high‑quality reference exploration video for each phantom. It achieves a mean translation error of 0.5 mm and increases frame‑wise localization coverage from 50.5 % to 86.1 % compared to standard Structure‑from‑Motion. The reconstructed trajectories reveal significant differences in navigation metrics between high‑ and low‑experience trainees, enabling objective skill assessment without external tracking equipment.
STATERA is a method that adapts a pretrained video backbone with mostly frozen weights and a lightweight temporal tubelet mixer to estimate the center-of-mass (CoM) of opaque, asymmetric rigid bodies from short monocular videos. It introduces the HiddenMass Benchmark, consisting of 50K simulated MuJoCo trajectories and a 63-sequence real-world test set with calibrated CoM ground truth. In simulation, STATERA reduces normalized CoM error from 41.7% to 25.2%, and in zero-shot sim-to-real transfer, its phase‑aware variant consistently predicts movement toward the true hidden offset, improving physics capture from 2.6% to 41.0%.