Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 20 of 309 for “"ImageNet"”.
-
Same data, same features: modern ImageNet-trained convolutional neural networks learn the same thing
Deep convolutional neural networks (CNNs) are the dominant technology in computer vision today. Much of the recent computer vision literature can be thought of as a competition to find the best architecture for vision within the deep convolutional framework. Despite all the effort invested in …
-
Developing TomoNet: A Deep Learning Feature Extractor for Medical Image Classification
… extractor pretrained with natural images from ImageNet (referred to as ImageNet-derived) with a feature extractor trained from scratch without any pre-existing feature extraction capabilities. Both feature extractors were trained to classify medical images into nineteen categories. The results …
-
Regularization, Uncertainty Estimation and Out of Distribution Detection in Convolutional Neural Networks
… of what the image consists of. For example an ImageNet-1K trained CNN can not say if the given image has no objects that it was trained on if it is provided with an image of a dinosaur (not an ImageNet category) or if the image has the main object cut out of it (context only). We approach these …
-
Constraining Robust Information Quantities Improves Adversarial Robustness
… against C&W and AutoAttack on CIFAR and Tiny-ImageNet datasets, both with and without additional synthetic data. The results show that our approach enhances the robust accuracy of DNNs by up to 2.66% on CIFAR datasets and 3.49% on Tiny-ImageNet against PGD attacks, and by 1.70% on CIFAR and …
-
Automated Segmentation of Head and Neck Cancer on Computed Tomography Imaging
… including out-of-domain pretraining using ImageNet and in-domain pretraining using the Jigsaw task. All in-domain pretraining, both using Jigsaw and the novel segmentation method created here, were performed on a composite pretraining dataset of over 618,000 CT images which was created by …
-
MSTOK: Προσαρμογή Διακριτοποίησης Εικόνων σε Πολυφασματικές Εικόνες Παρατήρησης της Γης
… την αρχικοποίηση από προεκπαιδευμένο μοντέλο στο ImageNet, αρχιτεκτονικές προσαρμογές και την κοινή εκπαίδευση σε ImageNet και TerraMesh. Η ποιότητα ανακατασκευής αξιολογείται σε 89 088 δείγματα επικύρωσης ανάλυσης 224 × 224, με πολυφασματικές μετρικές και αντιληπτικές μετρικές RGB. Μετά από μία …
-
DES: Delayed Eps-Shrinking for Faster Once-For-All Training
… across different datasets including CIFAR10/100, ImageNet-100, and ImageNet-1k on accuracy and cost. It achieves 1.83% higher ImageNet-1k top1 accuracy or the same accuracy with 1.3x x reduction in FLOPs and 2.5x drop in training cost (GPU*hrs).
-
Utilizing network features to detect erroneous inputs
… with an AUROC of 0.973 on CIFAR10, 0.957 on Tiny ImageNet, and 0.941 on ImageNet. I experimentally validate the findings across a diverse range of datasets, domains, and pre-trained models.
-
Exploring internally dense but externally sparse deep convolutional neural networks
… on benchmark tasks (CIFAR10, CIFAR100, and ImageNet) while keeping the network slim. Moreover, the researchers give out connection contributions by implementing a network damage experiment, which could indicate brief principles for hierarchical deep neural network design. The proposed method …
-
Whatls Left After Distillation? How Knowledge Transfer Impacts Fairness and Bias
… classification datasets, such as CIFAR-100, Tiny ImageNet and ImageNet, as many as 41% of the classes are statistically significantly affected by distillation when comparing class-wise accuracy (i.e. class bias) between a teacher/distilled student or distilled student/non-distilled student model. …
-
Autoregressive Modeling for Visual Understanding and Generation
… Transformer-Huge (ViT-H) encoder reaches 90.0% ImageNet top-1 accuracy. ARM aligns autoregressive pretraining with the directional computation of Mamba and develops MambaMLP blocks that support causal pretraining and multi-directional fine-tuning; ARM-H reaches 85.0% ImageNet accuracy. ARVideo …
-
MRL-AdANNS: Matryoshka Representation Learning for Web-Scale Adaptive Semantic Search
… tations, with benefits like up to 14× smaller ImageNet-1K embeddings and 14× speed-ups for large-scale retrieval. It extends seamlessly to web-scale datasets (ImageNet, JFT) across Vision (ResNet, ViT), Language (BERT), and V+L (ALIGN) modalities. In modern web-scale search systems, rigid …
-
Class Incremental Learning in Deep Neural Networks
… on common benchmarks using CIFAR10/100, CUB and ImageNet datasets. Results show that the proposed IGRE framework outperforms state-of-the-art methods on CIFAR-10, CIFAR-100, and the CUB datasets with 6-9\% improvement in accuracy and achieves comparable performance in large-scale ImageNet …
-
Learning Efficient Datatypes for Deep Neural Network Training and Inference
… between the nearest integer bitlengths. On ImageNet, BitPruning achieves average bitlength of 3.76 and 4.36 bits for ResNet18 and MobileNetV2, respectively, while staying within 0.5% of baseline accuracy. For training acceleration, Quantum Integer introduces a stochastic bitlength approach …
-
High-Resolution Queries, Low-Resolution Context: Scaling Vision Transformers with Asymmetric Spatial Reduction
… are conducted on CIFAR-100 at R = 2 and on ImageNet-1K at com pression rates of R = 2 and R = 4. On ImageNet-1K, C-SRA at R = 2 achieves 76.25% Top-1 accuracy, within 0.17 percentage points of MHA’s 76.42%, while using 3.3 million fewer parameters and improving throughput by 3.3% and peak …
-
A Discrete Wavelet Transform GAN for NonHomogeneous Dehazing
… in feature maps. To prevent over-fitting, ImageNet pre-trained Res2Net is adopted in the knowledge adaptation branch. Owing to the robust feature representations of ImageNet pre-training, the generalization ability of our network is improved dramatically. Finally, a patch-based …
-
On label-efficient computer vision : building fast and effective few-shot image classifiers
… achieve strong performance on Meta-Dataset, mini-ImageNet and tiered-ImageNet benchmarks. We further extend this approach to a transductive learning setting, proposing Transductive CNAPS. This transductive method combines a soft k-means parameter refinement procedure with a two-step task encoder …
-
Automatické označování obrázků
… descent. Je navrhnuté riešenie s hierarchiou pre ImageNet a tagovanie obrázkov atribútmi. Výpočetný model MapReduce je ukázaný pre učenie na veľkých dátových sadách. V poslednej časti je popísaná implementácia, experimentálne a testovacie výsledky.
-
Advances in Evaluation Metrics and Sampling Techniques for Generative Image Models
… human judgments by examining its sensitivity to ImageNet classes, and discussing its implications for generative model evaluation. We then propose an improved precision and recall metric that separately quantifies the fidelity and diversity of generated samples through explicit, non-parametric …
-
Enhancing Deep Learning Model Robustness: Insights from Out of Distribution Data Augmentation and an Innovative Image Compression Technique
… for convolutional neural networks trained on ImageNet. However, our evaluation of 39 augmentations challenges this belief, showing that an augmentation-induced in- crease in shape bias does not necessarily correlate with higher OOD robustness. Analyzing results, we identify biases in ImageNet …
Page 1 of 16