Paper compares GAN techniques for image generation and modification.
problem Improving image generation and modification techniques using GANs.
method Comparison of supervised and unsupervised GANs, addition of an encoder, use of Capsule Network as discriminator.
result Reconstruction and modification of images possible with GANs.
Method explains anomaly detection by generating normal modifications.
problem Complexity of deep learning methods in anomaly detection.
method Generates multiple alternative modifications for anomalies.
result High-quality semantic explanations provided for anomaly detection.
Automatic video modification to hide faces while maintaining pose, illumination, and expression.
problem Face de-identification in video to protect identities.
method A novel feed-forward encoder-decoder network architecture conditioned on facial image high-level representation.
result Fully automatic video modification at high frame rates with minimal distortion.
A stealthy framework injects faults into DNNs to misclassify images without affecting overall accuracy.
problem Vulnerability of deep neural networks to misclassification attacks.
method Fault sneaking attack using ADMM optimization with constraints on maintaining model accuracy and minimizing parameter modifications.
result The framework can inject multiple sneaking faults into DNNs without reducing overall accuracy.
We define an operation on homology B4 which we call an n-twist annulus modification. We give a new construction of smoothly slice knots and exotically slice knots via n-twist annulus modifications. As an application, we present a new example of a smoothly slice knot with non-slice derivatives. Such examples we…
Proposes a new method to generate unrestricted adversarial examples.
problem Generating unrestricted adversarial examples without norm constraints.
method Leveraging state-of-the-art generative models to manipulate image fine-grained aspects.
result Our adversarial images look indistinguishable from natural images and can bypass certified defenses.
A new method for effective VAE training using calibrated decoders.
problem Training VAEs requires hyperparameter tuning, leading to inefficiency.
method Calibrated decoders that learn uncertainty and automatically determine information retention.
result Calibrated decoders can simplify VAE training without heuristic modifications.
Improved VGG networks enhance image classification accuracy.
problem Enhancing image classification accuracy using modified VGG architectures.
method Two improved VGG architectures were created by freezing the first two blocks and applying different dilation rates in the last three blocks.
result Significant out-performance on image classification tasks on CIFAR-10 and CIFAR-100 datasets.
Enhances GANs by improving consistency regularization.
problem Improving the artifacts introduced by consistency regularization in GANs.
method Proposed modifications to consistency regularization to fix artifacts and improve performance.
result Significant improvement in FID scores on various GAN architectures.
Proposes a probabilistic method for generating semantically-aware adversarial examples.
problem Generating adversarial examples that are difficult for humans to detect while preserving semantics.
method Embeds subjective understanding of semantics as a distribution into adversarial example generation.
result Achieves higher success rates in circumventing adversarial defense mechanisms.
Tampering data during training can corrupt neural networks, leading to misclassification.
problem Adversarial attacks on neural networks during training.
method Creating a backdoor in neural networks by modifying training data.
result A single pixel modification to all images of a class in training data can corrupt neural networks, leading to misclassification.
Quickshift++ improves clustering stability and performance.
problem Improving initial seedings for clustering algorithms.
method Provably good initial seedings for Quick Shift clustering.
result Statistical consistency and strong clustering performance.
One pixel modification can make deep models unlearnable.
problem Protecting data from unauthorized training of deep neural networks.
method Perturbing only one pixel in each image to degrade model accuracy.
result Generated One-Pixel Shortcut (OPS) cannot be erased by adversarial training and strong augmentations.
RNA accelerates CNNs for image recognition.
problem Improving the optimization process of CNNs for image recognition.
method Regularized Nonlinear Acceleration (RNA) applied to neural networks.
result RNA improves the optimization process of CNNs slightly.
Generates high-quality images using sparse DCT representations.
problem Challenges in generating images due to high dimensionality.
method Transformers trained on sparse DCT block sequences.
result Competitive image generation quality with state-of-the-art methods.
New algorithms estimate matrix norms without matrix multiplication.
problem Estimating matrix norms efficiently in a matrix-free setting.
method Randomized algorithms based on Hutchinson's estimator modifications.
result Oracle complexity bounds for two-to-infinity and one-to-two norms.
SNS-GAN integrates class labels into generative models for images and time series.
problem Effective integration of class labels in generative models without network modifications.
method Embeds class conditions within the generator's noise space.
result Superior performance in time series generation compared to baseline models.
Physical patches can fool object detectors anywhere in an image.
problem Physical adversarial attacks on object detection systems.
method Demonstrated a patch that suppresses all detected objects regardless of its position.
result A patch can be placed anywhere in an image to suppress all detected objects, including those far away.
MCD automates counterfactual design searches for multi-modal tasks.
problem Designing for multi-objective goals and complex constraints.
method Model-agnostic counterfactual search method for multi-modal design modifications.
result MCD streamlines and automates counterfactual search, recommending effective design modifications.
Enhances facial emotion recognition with gradient and Laplacian images.
problem Improving the performance of facial emotion recognition systems.
method Proposes using gradient and Laplacian of input images with a CNN.
result Enhances FER systems by 3 to 5%.
A new tensor network method for image classification reduces computation cost.
problem Efficiently classifying images in high-dimensional spaces.
method Proposes a multi-layered tensor network (MLTN) that performs one MPS operation per layer, reducing computation cost.
result Reduces computation cost without degrading performance.
LMConv improves autoregressive models for image generation and completion.
problem Limited generation order in autoregressive models restricts their applicability.
method Introduces LMConv, a modified 2D convolution that allows arbitrary masks to be applied to weights.
result LMConv achieves improved performance on image density estimation and coherent completions.
Adversarial examples and noisy images share a common cause.
problem Improving machine learning models' robustness to adversarial attacks and random noise.
method Empirical and theoretical analysis of adversarial examples and corrupted images.
result Adversarial robustness and corruption robustness are closely related.
Plug-and-play multimodal controller improves class-conditional image generation.
problem Generating class-conditional images from user-specified labels.
method Introduces a `multimodal controller` to generate multimodal data without additional learning parameters.
result Multimodal controlled generative models produce higher quality class-conditional images and novel modalities.
A dynamic ResNet model learns different routes for images from different classes.
problem Fixed structure in ResNet-like architectures limits their adaptability to diverse inputs.
method Develops a ResNet-based model that dynamically selects Computational Units for each input image.
result Achieves better results on CIFAR-10 test set compared to the original ResNet-38 architecture.
Proposes a technique to interpret deep learning models by generating counterfactual inputs.
problem Understanding and explaining the decisions made by deep neural networks.
method Uses a generative model to edit input images and generate counterfactual scenarios for model interpretation.
result Demonstrates the effectiveness of the introspection approach on MNIST and CelebA datasets.
Improves conditional image generation quality on ImageNet dataset.
problem Improving conditional image generation quality on ImageNet dataset.
method Projection-based discriminator modification in GANs.
result Significantly improved image generation quality on ILSVRC2012 dataset.
Quantum machine learns to clean up blurry images.
problem Cleaning up blurry images using quantum computing.
method Uses Boltzmann machines, QUBO, and quantum annealing to balance image quality and noise.
result Quantum method produces cleaner images than noisy originals on average.
BigGAN achieves state-of-the-art image synthesis on ImageNet.
problem Generating high-fidelity images from complex datasets like ImageNet.
method Trained a large-scale GAN with orthogonal regularization and a truncation trick.
result Improved Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4.
We address the following problem: given two smooth densities on a manifold, find an optimal diffeomorphism that transforms one density into the other. Our framework builds on connections between the Fisher-Rao information metric on the space of probability densities and right-invariant metrics on the infinite-dimension…
Transforms improve CNNs' invariance to image transformations.
problem Current CNN models lack robustness to spatial transformations.
method Randomly transform feature maps during training to learn invariant representations.
result Significant improvements on benchmark tasks, including image recognition and retrieval.
DreamFusion uses text-to-image diffusion models to create 3D images efficiently.
problem Lack of large-scale 3D datasets and efficient architectures for 3D synthesis.
method Adapting a 2D diffusion model to 3D synthesis using a loss based on probability density distillation.
result A 3D model can be optimized from a 2D diffusion model, allowing for text-to-3D synthesis.
Unified framework detects natural and adversarial errors in image classifications.
problem Detecting both unintentional and intentional errors in image classifications.
method Detects errors using invariance to image transformations.
result Our approach surpasses previous methods by a large margin.
Improved image restoration using frequency-guided sampling.
problem Restoring high-quality images from degraded observations with known degradation processes.
method Proposed a frequency-guided sampling approach for diffusion-based image restoration, incorporating a time-varying low-pass filter.
result Significantly improved performance on challenging image restoration tasks, including motion deblurring and image dehazing.
Deep learning transforms time series into images for anomaly detection in industrial assets.
problem Detecting anomalies in time series data from industrial assets.
method Transforming time series data into image-like representations and using them as inputs for deep learning models.
result Some encodings provide competitive results for anomaly detection in industrial asset monitoring.
Proposes methods to improve hierarchical classification accuracy by flattening inconsistent nodes.
problem Error propagation in top-down hierarchical classification due to inconsistent nodes.
method Data-driven approaches for identifying and flattening inconsistent nodes.
result Improves classification performance by up to 7% in Macro-F1 score.
Study analyzes overfitting in neural nets under class imbalance, proposing new loss functions.
problem Overfitting and class imbalance in neural nets, particularly in image segmentation.
method Analyzes logits distribution, derives asymmetric loss functions and regularizers.
result Proposed loss functions improve segmentation performance in brain tumor core.
Improves model accuracy in medical imaging with small datasets using transfer learning.
problem Challenges in training neural networks on small medical imaging datasets.
method Comparison of current techniques, proposing one cycle training, discriminative learning rates, gradual freezing, and parameter modification.
result Transfer learning is crucial for small datasets, especially when images from the same part of the body are available.
Simple image augmentation boosts deep RL from pixels.
problem Training deep RL models from raw pixel inputs is challenging.
method Applying standard computer vision data augmentation to reinforcement learning.
result Our method significantly improves SAC's performance on DeepMind control suite.
Turbo-DDCM speeds up zero-shot image compression.
problem Slowness and high computational demand in zero-shot diffusion-based compression.
method Modified DDCM framework with Turbo-DDCM, combining noise vectors and improved encoding.
result Turbo-DDCM achieves faster compression than state-of-the-art methods.
Develops new methods to create imperceptible image changes that fool classifiers.
problem Improving the robustness of image classifiers by creating subtle changes undetectable to humans.
method Two methods: Edge-Aware and Color-Aware, designed to reduce detectability of image perturbations.
result Demonstrated that the new methods effectively cause misclassification and are computationally efficient.
Improved autoregressive models generate higher quality images and are more robust to noise.
problem Generating high-quality images from autoregressive models.
method Noise conditional maximum likelihood estimation (MLE) with score-based sampling.
result Models trained with noise conditional MLE achieve better test likelihoods and generate higher quality images.
This work improves testing of machine learning model modifications using novel statistical methods.
problem Overfitting and conservative Bonferroni correction when testing multiple model modifications.
method Introduces alpha-recycling and SRGPs to control error rate and approve more beneficial modifications.
result Novel statistical methods approve a higher number of beneficial modifications than previous approaches.
Machine learning models in physical world are vulnerable to subtle adversarial examples.
problem Vulnerability of machine learning models to adversarial examples in physical world.
method Demonstrated vulnerability by feeding adversarial images from cell-phone camera to an ImageNet Inception classifier.
result A large fraction of adversarial examples are classified incorrectly even through a camera.
PaRCE estimates model confidence for CNNs across various uncertainties.
problem Limited holistic approach to estimating perception model confidence in CNNs.
method Probabilistic and reconstruction-based competency estimation.
result PaRCE best distinguishes between various types of samples and regions.
Study identifies and mitigates causes of image misclassifications in CNN models.
problem Improving model interpretability and accuracy in image classification.
method Trained six CNN architectures on CIFAR-10, used conditional confusion matrices and misclassification networks to identify morphological similarity and non-essential information interference as causes of misclassification. Developed a method to reduce misclassifications by erasing pixels within top 5% saliency map bounding boxes.
result Identified two causes of misclassification: morphological similarity and non-essential information interference, and developed a method to reduce misclassifications caused by the latter.
SeqAttnGAN generates interactive images based on multi-turn text descriptions.
problem Interactive image editing with multi-turn textual commands.
method SeqAttnGAN uses a neural state tracker and GAN framework for sequential image generation and refinement.
result SeqAttnGAN outperforms state-of-the-art models on interactive image editing tasks.
Adversarial attacks can fool self-driving cars, changing steering predictions.
problem Security of deep neural networks in self-driving cars.
method Demonstrated adversarial attacks on steering angle prediction model.
result Small image modifications can mislead self-driving car predictions.