The paper presents a two‑stage approach to street‑panorama semantic segmentation using Vision Transformer ensembles. In the first stage, nine pretrained models are compared under equal compute budgets, and the best two are trained with three random seeds, achieving 60.95% mIoU on validation and 57.08% on the hidden test leaderboard. The second stage explores augmentations, larger inputs, and additional architectures, culminating in an ensemble of four models that scores 64.53% mIoU on validation and 61.54% mIoU (71.93% F1) on the hidden test leaderboard, while the study documents all runs, costs, and releases the code and models.
By Yunus Serhat B{\i}\c{c}ak\c{c}{\i}
arXiv:2609.01584v1 Announce Type: new
Abstract: Vehicle attribute analysis is a key component of Intelligent Transportation Systems (ITS), supporting applications such as vehicle identification, traf...
By Sergio M. Silva Jr., Otavio T. Remer, Gabriel E. Lima, Lucas Wojcik, Rayson Laroca, David Menotti
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.
By Tamoghna Chakraborty, Md Nurul Absur, Sourya Saha, Saptarshi Debroy
arXiv:2607. 13234v1 Announce Type: cross Abstract: Deepfake detectors that achieve near-perfect scores on academic benchmarks collapse on real-world content: recent in-the-wild evaluations report AUC drops of 45-50% for state-of-the-art open-source models.
By Ken Jon Miyachi, Dylan Uys
arXiv:2606. 17615v1 Announce Type: cross Abstract: Estimating human proficiency from video is a key challenge for automated skill assessment, with applications in sports coaching, music pedagogy, surgical training, and workplace learning.
By Edoardo Bianchi, Antonio Liotta
PairCLIP‑SWA is a single‑backbone deepfake detector that fine‑tunes an OpenCLIP ViT‑L/14 visual tower using a class‑balanced corpus created by CLIP‑similarity‑guided subsampling and then applies stochastic weight averaging over the fine‑tuning trajectory. Trained on the IEEE SP Cup 2025 (DF‑Wild‑Cup) benchmark, the weight‑averaged model achieves 97.04 % accuracy, 98.64 % AU‑ROC and a 4.07 % equal error rate—outperforming the best published three‑backbone ensemble. The study shows that averaging stabilizes validation performance, prevents misranking on disjoint test sets, and that CLIP pre‑training, rather than ensembling, is the key driver of accuracy.
By Md. Nakib, Sudipto Sarkar, A. J. A. Jubayer Talukder, Mohammad Ariful Haque
TAPe+ML v3 is a compact computer vision system that uses a structured representation called TAPe to encode relationships among perceptual elements before recognition. The system employs a shared TAPe representation and a modular recognition architecture for tasks such as image classification, object detection, and instance segmentation, achieving strong performance with fewer than 100,000 parameters. Experiments show high mAP scores on COCO, strong classification accuracy on Imagenette and ImageNet-Real, and demonstrate benefits in video scene detection and distribution‑shift adaptation in an industrial pilot.
By Sergey Kurinov (Comexp Research Lab, TAPe + ML Project, Nizhniy Novgorod, Russia), Alexey Upatov (Comexp Research Lab, TAPe + ML Project, Nizhniy Novgorod, Russia)
arXiv:2609.05762v1 Announce Type: cross
Abstract: UAV image collections contain spatially and temporally related frames, yet semantic-segmentation benchmarks commonly split them at image level. Such...
By Tarek Rahman, Nazim-E-Alam, Md Kishor Morol, Jannatun Noor
LeVJEPA is a video encoder that eliminates the need for architectural asymmetries, exponential-moving-average target encoders, stop-gradients, and capacity-limited predictors used in prior self‑supervised methods. It trains a single encoder with an invariance loss over global and local views, regularized by SIGReg to prevent collapse, and achieves strong performance with far less pretraining compute. The approach also allows block‑causal attention, making temporal ordering a property of the encoder itself, and matches or surpasses state‑of‑the‑art baselines on both appearance‑centric and motion‑centric benchmarks.
By Lukas Kuhn, Lucas Maes, Giuseppe Serra, Quentin Le Lidec, Yann LeCun, Randall Balestriero, Florian Buettner
arXiv:2606. 08123v2 Announce Type: replace-cross Abstract: Model selection for safety-relevant visual recognition is often based on clean aggregate performance, although robustness, transfer, embedded latency, and explanation faithfulness may produce different preferences.
By Ruben Dario Florez-Zela
arXiv:2608. 07579v1 Announce Type: cross Abstract: The AI City Challenge 2026 Track 1 evaluates multi-camera 3D perception in large indoor warehouses under a synthetic-to-real (Sim2Real) setting; depth is available only for training and validation, so inference is RGB-only.
By Abdullah Naeem, Anav Katwal, Ayon Dey, Noman Khan, Md Tamjidul Hoque
arXiv:2608. 11770v1 Announce Type: cross Abstract: Edge-deployed vision systems in target recognition, surveillance, autonomous vehicles, and drone domains require hierarchical inference pipelines where a detection model identifies objects of interest and downstream classifiers provide fine-grained attribute analysis.
By Vaishnav Raju