PyTorchKR Papers

“Image Classification”

총 18,385편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2403.18198공식 코드 · Papers with Code

Generative Medical Segmentation

Rapid advancements in medical image segmentation performance have been significantly driven by the development of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). These models follow the discriminative pixel-wise classification learning paradigm and often have limited ability to generalize across diverse medical imaging datasets. In this manuscript, we introduce Generative Medical Segmentation (GMS), a novel approach leveraging a generative model to perform image segmentation. Concretely, GMS employs a robust pre-trained vision foundation model to extract latent representations for images and corresponding ground truth masks, followed by a model that learns a mapping function from the image to the mask in the latent space. Once trained, the model generates an estimated segmentation mask using the pre-trained vision foundation model to decode the predicted latent representation back into the image space. The design of GMS leads to fewer trainable parameters in the model which reduces the risk of overfitting and enhances its generalization capability. Our experimental analysis across five public datasets in different medical imaging domains demonstrates GMS outperforms existing discriminative and generative segmentation models. Furthermore, GMS is able to generalize well across datasets from different centers within the same imaging modality. Our experiments suggest GMS offers a scalable and effective solution for medical image segmentation. GMS implementation and trained model weights are available at https://github.com/King-HAW/GMS.

arXiv:2602.23783v5cs.CV

Diffusion Probe: CNN 프로브를 활용한 생성 이미지 결과 예측

Diffusion Probe: Generated Image Result Prediction Using CNN Probes

텍스트-이미지(Text-to-image, T2I) 확산 모델은 조기에 품질을 평가하는 효율적인 메커니즘이 부족하여 프롬프트 반복(iteration), 에이전트(agent) 기반 생성, flow-grpo와 같이 여러 번 생성을 거치는 시나리오에서 많은 비용이 드는 시행착오를 초래합니다. 본 연구에서는 초기 확산 크로스 어텐션(cross-attention) 분포와 최종 이미지 품질 사이에 강한 상관관계가 있음을 밝힙니다. 이러한 발견을 바탕으로, 내부 크로스 어텐션 맵을 예측 신호로 활용하는 프레임워크인 Diffusion Probe를 제안합니다. 초기 노이즈 제거(denoising) 단계에서 추출한 초기 크로스 어텐션의 통계적 특성을 최종 이미지의 전반적인 품질로 매핑하는 경량 예측기를 설계합니다. 이를 통해 전체 합성이 완료되기 훨씬 전에 다양한 평가 지표에 걸쳐 이미지 품질을 정확하게 예측할 수 있습니다. 본 연구는 다양한 설정에서 Diffusion Probe를 검증합니다. 여러 T2I 모델, 초기 노이즈 제거 구간, 해상도 및 품질 지표 전반에 걸쳐 강력한 상관관계(PCC > 0.7)와 높은 분류(classification) 성능(AUC-ROC > 0.9)을 달성합니다. 이러한 신뢰성은 실질적인 이점으로 이어집니다. 프롬프트 최적화, 시드 선택, 가속화된 강화학습(RL) 학습(training)과 같은 워크플로우에서 품질 기반의 조기 의사결정을 가능하게 함으로써, 프로브는 보다 표적화된 샘플링을 지원하고 잠재력이 낮은 생성에 대한 연산을 방지합니다. 이는 최종 출력 품질을 개선하면서도 연산 오버헤드를 줄여줍니다. Diffusion Probe는 모델에 구애받지 않고 효율적이며 광범위하게 적용 가능하여, 조기 품질 예측을 통해 T2I 생성 효율성을 개선하는 실용적인 솔루션을 제공합니다.

arXiv:2611.20841v1cs.CV

비전-언어 모델을 이용한 위상 감응형 광 시간 영역 반사 측정의 이벤트 분류

Vision-Language Model-Based Phase-OTDR Event Classification

위상 감응형 광 시간 영역 반사 측정(Phase-OTDR)은 분산 진동 감지를 가능하게 한다. 다만 신뢰할 수 있는 이벤트 분류에는 전통적으로 많은 주석 데이터가 필요한 완전 지도학습 모델을 사용해 왔다. 본 연구는 퓨샷 학습을 통해 비전-언어 모델(VLM)을 Phase-OTDR 이벤트 인식에 적용할 수 있는지 살핀다. 원시 표본을 단시간 푸리에 변환(STFT)으로 데시벨 크기의 스펙트로그램 모자이크로 바꾼다. 색상을 매핑하고 크기를 조정해 RGB 이미지로 만든 뒤 대조 언어-이미지 사전학습(CLIP) 이미지 인코더로 처리한다. 퓨샷 적응에서는 Tip-Adapter-F의 캐시 키만 최적화하고 두 인코더는 그대로 둔다. 누출을 통제하기 위해 학습 쪽의 지원·검증 풀에 클래스당 이미지 512장을 두고, 고정 평가 세트에 클래스당 120장을 별도로 둔다. 구성 선택에는 평가 세트를 쓰지 않았고 지원 표본을 다르게 뽑아 20회 시험했다. 선택된 구성은 OpenAI 가중치의 ViT-L-14 비전 트랜스포머 인코더와 클래스당 128샷이었다. 별도 평가 세트에서 정확도 96.81 ± 0.42%, 매크로 평균 F1 96.79 ± 0.42%를 달성했다. 여섯 클래스에 걸쳐 레이블이 있는 지원 이미지 96장만 적응에 사용하는 16샷 설정에서도 정확도가 90.17 ± 0.81%였다. Tip-Adapter-F는 제로샷 CLIP과 학습 없는 표준 Tip-Adapter를 모두 뚜렷하게 앞섰다. 프롬프트와 백본을 분석한 결과, 선택한 설정에서 성능은 비전 트랜스포머 인코더를 바꿔도 안정적이었고 프롬프트 문구에 민감하지 않았다. CLIP의 시각 특징은 다른 과제로 전이할 수 있는 구조를 가지며, 가벼운 지원 캐시를 더해 적은 레이블로 Phase-OTDR 이벤트를 분류할 수 있다. 후속 연구를 위해 코드와 이미지 데이터셋을 https://github.com/miralab-ai/vlm-phase-otdr 에 공개한다.

arXiv:2505.09252공식 코드 · Papers with Code

Zero-Shot Multi-modal Large Language Model v.s. Supervised Deep Learning: A Comparative Study on CT-Based Intracranial Hemorrhage Subtyping

Introduction: Timely identification of intracranial hemorrhage (ICH) subtypes on non-contrast computed tomography is critical for prognosis prediction and therapeutic decision-making, yet remains challenging due to low contrast and blurring boundaries. This study evaluates the performance of zero-shot multi-modal large language models (MLLMs) compared to traditional deep learning methods in ICH binary classification and subtyping. Methods: We utilized a dataset provided by RSNA, comprising 192 NCCT volumes. The study compares various MLLMs, including GPT-4o, Gemini 2.0 Flash, and Claude 3.5 Sonnet V2, with conventional deep learning models, including ResNet50 and Vision Transformer. Carefully crafted prompts were used to guide MLLMs in tasks such as ICH presence, subtype classification, localization, and volume estimation. Results: The results indicate that in the ICH binary classification task, traditional deep learning models outperform MLLMs comprehensively. For subtype classification, MLLMs also exhibit inferior performance compared to traditional deep learning models, with Gemini 2.0 Flash achieving an macro-averaged precision of 0.41 and a macro-averaged F1 score of 0.31. Conclusion: While MLLMs excel in interactive capabilities, their overall accuracy in ICH subtyping is inferior to deep networks. However, MLLMs enhance interpretability through language interactions, indicating potential in medical imaging analysis. Future efforts will focus on model refinement and developing more precise MLLMs to improve performance in three-dimensional medical image processing.

arXiv:1906.03808

A Closed-Form Learned Pooling for Deep Classification Networks

In modern computer vision tasks, convolutional neural networks (CNNs) are indispensable for image classification tasks due to their efficiency and effectiveness. Part of their superiority compared to other architectures, comes from the fact that a single, local filter is shared across the entire image. However, there are scenarios where we may need to treat spatial locations in non-uniform manner. We see this in nature when considering how humans have evolved foveation to process different areas in their field of vision with varying levels of detail. In this paper we propose a way to enable CNNs to learn different pooling weights for each pixel location. We do so by introducing an extended definition of a pooling operator. This operator can learn a strict super-set of what can be learned by average pooling or convolutions. It has the benefit of being shared across feature maps and can be encouraged to be local or diffuse depending on the data. We show that for fixed network weights, our pooling operator can be computed in closed-form by spectral decomposition of matrices associated with class separability. Through experiments, we show that this operator benefits generalization for ResNets and CNNs on the CIFAR-10, CIFAR-100 and SVHN datasets and improves robustness to geometric corruptions and perturbations on the CIFAR-10-C and CIFAR-10-P test sets.

arXiv:2306.07490공식 코드 · Papers with Code

Top-Down Framework for Weakly-supervised Grounded Image Captioning

Weakly-supervised grounded image captioning (WSGIC) aims to generate the caption and ground (localize) predicted object words in the input image without using bounding box supervision. Recent two-stage solutions mostly apply a bottom-up pipeline: (1) encode the input image into multiple region features using an object detector; (2) leverage region features for captioning and grounding. However, utilizing independent proposals produced by object detectors tends to make the subsequent grounded captioner overfitted in finding the correct object words, overlooking the relation between objects, and selecting incompatible proposal regions for grounding. To address these issues, we propose a one-stage weakly-supervised grounded captioner that directly takes the RGB image as input to perform captioning and grounding at the top-down image level. Specifically, we encode the image into visual token representations and propose a Recurrent Grounding Module (RGM) in the decoder to obtain precise Visual Language Attention Maps (VLAMs), which recognize the spatial locations of the objects. In addition, we explicitly inject a relation module into our one-stage framework to encourage relation understanding through multi-label classification. This relation semantics served as contextual information facilitating the prediction of relation and object words in the caption. We observe that the relation semantic not only assists the grounded captioner in generating a more accurate caption but also improves the grounding performance. We validate the effectiveness of our proposed method on two challenging datasets (Flick30k Entities captioning and MSCOCO captioning). The experimental results demonstrate that our method achieves state-of-the-art grounding performance.

arXiv:2407.01530공식 코드 · Papers with Code

xLSTM-UNet can be an Effective 2D & 3D Medical Image Segmentation Backbone with Vision-LSTM (ViL) better than its Mamba Counterpart

Convolutional Neural Networks (CNNs) and Vision Transformers (ViT) have been pivotal in biomedical image segmentation, yet their ability to manage long-range dependencies remains constrained by inherent locality and computational overhead. To overcome these challenges, in this technical report, we first propose xLSTM-UNet, a UNet structured deep learning neural network that leverages Vision-LSTM (xLSTM) as its backbone for medical image segmentation. xLSTM is a recently proposed as the successor of Long Short-Term Memory (LSTM) networks and have demonstrated superior performance compared to Transformers and State Space Models (SSMs) like Mamba in Neural Language Processing (NLP) and image classification (as demonstrated in Vision-LSTM, or ViL implementation). Here, xLSTM-UNet we designed extend the success in biomedical image segmentation domain. By integrating the local feature extraction strengths of convolutional layers with the long-range dependency capturing abilities of xLSTM, xLSTM-UNet offers a robust solution for comprehensive image analysis. We validate the efficacy of xLSTM-UNet through experiments. Our findings demonstrate that xLSTM-UNet consistently surpasses the performance of leading CNN-based, Transformer-based, and Mamba-based segmentation networks in multiple datasets in biomedical segmentation including organs in abdomen MRI, instruments in endoscopic images, and cells in microscopic images. With comprehensive experiments performed, this technical report highlights the potential of xLSTM-based architectures in advancing biomedical image analysis in both 2D and 3D. The code, models, and datasets are publicly available at http://tianrun-chen.github.io/xLSTM-UNet/

arXiv:2611.11893v1cs.CV

건물 단위 빈집 평가를 위한 스트리트뷰 영상 처리 파이프라인 개발

Developing a Street View Imagery Processing Pipeline for Building-Level Vacant House Assessment

스트리트뷰 영상(SVI)은 도시 환경을 원격으로 점검하는 데 널리 사용된다. 자동화된 평가는 대개 각 영상이 대상 건물을 명확히 보여 준다고 가정한다. 실제로는 영상에 과업 관련 증거가 없거나 이웃 주택이 찍혀 있을 수 있다. 이를 확인하지 않으면 모델의 점수만으로는 올바른 건물을 판단했는지 알 수 없다. 우리는 영상 적합성과 대상 식별을 명시적이고 검증 가능하게 만드는 재현 가능한 SVI 처리 파이프라인을 제시한다. 이를 대한민국의 빈집 평가에 적용하였으며, 단독주택 필지 1,609개의 영상 10,311장을 사용하였다. 과업 적합성 필터링은 영상 구성과 행정구역에 따라 기준을 조정하며, 이로써 지역 간 포괄 범위를 고르게 유지한다. 단일 균일 필터와 같은 수의 영상을 유지하면서도 우리의 필터링은 사용 가능한 영상을 가진 필지를 101개 더 남겼다. 대상 주택 정렬은 카메라 메타데이터와 필지 경계로부터 해당 필지의 주택을 식별하며, 건물 외곽선이나 추가 학습이 필요하지 않다. 정렬은 영상의 85.6%에서 사람의 라벨과 일치하였으며, 가장 좋은 단순 규칙은 79.8%였다. 빈집 분류에서 필터링과 정렬은 AUROC를 0.684에서 0.719로 높였으나, 같은 영상에서 다른 주택을 표시해도 여전히 0.709를 보였다. Grad-CAM은 모델이 대상 주택 밖의 영역을 자주 본다는 것을 보여 주었다. 따라서 영상 적합성, 대상 식별, 모델 근거는 서로 별개의 문제이며, 점수만으로는 해결되지 않는다. 어떤 영상이 사용되었고 어떤 주택이 선택되었는지를 기록하면 검토자가 각 평가를 검증할 수 있고 사람의 가상 점검을 뒷받침할 수 있다.

arXiv:1809.04184

Searching for Efficient Multi-Scale Architectures for Dense Image Prediction

The design of neural network architectures is an important component for achieving state-of-the-art performance with machine learning systems across a broad array of tasks. Much work has endeavored to design and build architectures automatically through clever construction of a search space paired with simple learning algorithms. Recent progress has demonstrated that such meta-learning methods may exceed scalable human-invented architectures on image classification tasks. An open question is the degree to which such methods may generalize to new domains. In this work we explore the construction of meta-learning techniques for dense image prediction focused on the tasks of scene parsing, person-part segmentation, and semantic image segmentation. Constructing viable search spaces in this domain is challenging because of the multi-scale representation of visual information and the necessity to operate on high resolution imagery. Based on a survey of techniques in dense image prediction, we construct a recursive search space and demonstrate that even with efficient random search, we can identify architectures that outperform human-invented architectures and achieve state-of-the-art performance on three dense prediction tasks including 82.7\% on Cityscapes (street scene parsing), 71.3\% on PASCAL-Person-Part (person-part segmentation), and 87.9\% on PASCAL VOC 2012 (semantic image segmentation). Additionally, the resulting architecture is more computationally efficient, requiring half the parameters and half the computational cost as previous state of the art systems.

arXiv:2502.18104공식 코드 · Papers with Code

PromptMID: Modal Invariant Descriptors Based on Diffusion and Vision Foundation Models for Optical-SAR Image Matching

The ideal goal of image matching is to achieve stable and efficient performance in unseen domains. However, many existing learning-based optical-SAR image matching methods, despite their effectiveness in specific scenarios, exhibit limited generalization and struggle to adapt to practical applications. Repeatedly training or fine-tuning matching models to address domain differences is not only not elegant enough but also introduces additional computational overhead and data production costs. In recent years, general foundation models have shown great potential for enhancing generalization. However, the disparity in visual domains between natural and remote sensing images poses challenges for their direct application. Therefore, effectively leveraging foundation models to improve the generalization of optical-SAR image matching remains challenge. To address the above challenges, we propose PromptMID, a novel approach that constructs modality-invariant descriptors using text prompts based on land use classification as priors information for optical and SAR image matching. PromptMID extracts multi-scale modality-invariant features by leveraging pre-trained diffusion models and visual foundation models (VFMs), while specially designed feature aggregation modules effectively fuse features across different granularities. Extensive experiments on optical-SAR image datasets from four diverse regions demonstrate that PromptMID outperforms state-of-the-art matching methods, achieving superior results in both seen and unseen domains and exhibiting strong cross-domain generalization capabilities. The source code will be made publicly available https://github.com/HanNieWHU/PromptMID.

arXiv:2505.15644공식 코드 · Papers with Code

FragFake: A Dataset for Fine-Grained Detection of Edited Images with Vision Language Models

Fine-grained edited image detection of localized edits in images is crucial for assessing content authenticity, especially given that modern diffusion models and image editing methods can produce highly realistic manipulations. However, this domain faces three challenges: (1) Binary classifiers yield only a global real-or-fake label without providing localization; (2) Traditional computer vision methods often rely on costly pixel-level annotations; and (3) No large-scale, high-quality dataset exists for modern image-editing detection techniques. To address these gaps, we develop an automated data-generation pipeline to create FragFake, the first dedicated benchmark dataset for edited image detection, which includes high-quality images from diverse editing models and a wide variety of edited objects. Based on FragFake, we utilize Vision Language Models (VLMs) for the first time in the task of edited image classification and edited region localization. Experimental results show that fine-tuned VLMs achieve higher average Object Precision across all datasets, significantly outperforming pretrained models. We further conduct ablation and transferability analyses to evaluate the detectors across various configurations and editing scenarios. To the best of our knowledge, this work is the first to reformulate localized image edit detection as a vision-language understanding task, establishing a new paradigm for the field. We anticipate that this work will establish a solid foundation to facilitate and inspire subsequent research endeavors in the domain of multimodal content authenticity.

arXiv:2611.05192v1cs.CV

자연 영상에서 의료 스캔으로: 의료 영상 생성을 위한 확산 모델의 전이 가능성

From Natural Images to Medical Scans: Transferability of Diffusion Models for Medical Image Generation

자연 영상으로부터의 전이 학습은 지도 학습 기반 의료 영상 분석에서는 표준이며 자기 지도 학습에서도 점차 채택되고 있으나, 생성 모델링, 특히 확산 디노이저에서의 역할은 아직 충분히 탐구되지 않았다. 의료 확산 모델은 일반적으로 디노이저를 처음부터 훈련하며, 자연 영상 전이에 대한 체계적인 평가는 없다. 본 연구는 잠재 및 픽셀 패러다임, 모델 크기(131M-952M), 두 개의 방사선 데이터셋(RadImageNet, BraTS)에 걸쳐 이를 분석하고, 전이의 어떤 측면이 생성을 뒷받침하는지 묻는다. 결과는 다음과 같다. i) 자연 영상 초기화는 생성을 일관되게 향상시키며, 토크나이저가 사전 지식의 상당 부분을 제공하는 잠재 공간에서는 상대 gFID가 14% 개선되는 데 비해 픽셀 공간에서는 21-34% 개선된다. gFID 이득은 임상적 충실도의 대리 지표인 하위 병리 분류 성능 및 라디오믹 특징 FD와 동일한 경향을 보인다. ii) 이점은 의미론이 아닌 공간 구조를 반영한다. 의미론적 프로빙은 생성 품질을 예측하지 못하는 반면, 디노이저 내부 표현에 맞게 조정한 공간 구조 지표는 $|r|=0.85$로 gFID를 추적하며, 이는 공간적 조직이 사전 훈련의 효과가 나타나는 경로임을 시사한다. iii) 혼합 도메인 디노이저 훈련은 자연 영상과 의료 영상이 후반 블록까지 유사한 계층별 표현을 공유함을 보여 주며, 의료 훈련에 자연 영상을 섞으면 충실도가 향상된다. iv) 전이 이점은 자연 영상 초기화에 결부되어 있으며, 이는 도메인 내 의료 사전 훈련과 여러 재초기화 방식보다 우수하다. 전반적으로, 본 연구의 설정에서 자연 영상 디노이저는 의료 영상 디노이저로 긍정적으로 전이된다. 코드는 https://github.com/CAIROLab-Bern/rad-dit 에서 이용할 수 있다.

arXiv:1604.04326

Improving the Robustness of Deep Neural Networks via Stability Training

In this paper we address the issue of output instability of deep neural networks: small perturbations in the visual input can significantly distort the feature embeddings and output of a neural network. Such instability affects many deep architectures with state-of-the-art performance on a wide range of computer vision tasks. We present a general stability training method to stabilize deep networks against small input distortions that result from various types of common image processing, such as compression, rescaling, and cropping. We validate our method by stabilizing the state-of-the-art Inception architecture against these types of distortions. In addition, we demonstrate that our stabilized model gives robust state-of-the-art performance on large-scale near-duplicate detection, similar-image ranking, and classification on noisy datasets.

arXiv:1310.0316

Classifying Traffic Scenes Using The GIST Image Descriptor

This paper investigates classification of traffic scenes in a very low bandwidth scenario, where an image should be coded by a small number of features. We introduce a novel dataset, called the FM1 dataset, consisting of 5615 images of eight different traffic scenes: open highway, open road, settlement, tunnel, tunnel exit, toll booth, heavy traffic and the overpass. We evaluate the suitability of the GIST descriptor as a representation of these images, first by exploring the descriptor space using PCA and k-means clustering, and then by using an SVM classifier and recording its 10-fold cross-validation performance on the introduced FM1 dataset. The obtained recognition rates are very encouraging, indicating that the use of the GIST descriptor alone could be sufficiently descriptive even when very high performance is required.

arXiv:1201.05947

Examplers based image fusion features for face recognition

Examplers of a face are formed from multiple gallery images of a person and are used in the process of classification of a test image. We incorporate such examplers in forming a biologically inspired local binary decisions on similarity based face recognition method. As opposed to single model approaches such as face averages the exampler based approach results in higher recognition accu- racies and stability. Using multiple training samples per person, the method shows the following recognition accuracies: 99.0% on AR, 99.5% on FERET, 99.5% on ORL, 99.3% on EYALE, 100.0% on YALE and 100.0% on CALTECH face databases. In addition to face recognition, the method also detects the natural variability in the face images which can find application in automatic tagging of face images.

arXiv:1908.05730

Skin Lesion Segmentation and Classification for ISIC 2018 by Combining Deep CNN and Handcrafted Features

This short report describes our submission to the ISIC 2018 Challenge in Skin Lesion Analysis Towards Melanoma Detection for Task1 and Task 3. This work has been accomplished by a team of researchers at the University of Dayton Signal and Image Processing Lab. Our proposed approach is computationally efficient are combines information from both deep learning and handcrafted features. For Task3, we form a new type of image features, called hybrid features, which has stronger discrimination ability than single method features. These features are utilized as inputs to a decision-making model that is based on a multiclass Support Vector Machine (SVM) classifier. The proposed technique is evaluated on online validation databases. Our score was 0.841 with SVM classifier on the validation dataset.

arXiv:2110.06486

Understanding of Emotion Perception from Art

Computational modeling of the emotions evoked by art in humans is a challenging problem because of the subjective and nuanced nature of art and affective signals. In this paper, we consider the above-mentioned problem of understanding emotions evoked in viewers by artwork using both text and visual modalities. Specifically, we analyze images and the accompanying text captions from the viewers expressing emotions as a multimodal classification task. Our results show that single-stream multimodal transformer-based models like MMBT and VisualBERT perform better compared to both image-only models and dual-stream multimodal models having separate pathways for text and image modalities. We also observe improvements in performance for extreme positive and negative emotion classes, when a single-stream model like MMBT is compared with a text-only transformer model like BERT.

arXiv:2209.12459

Ablation Path Saliency

Various types of saliency methods have been proposed for explaining black-box classification. In image applications, this means highlighting the part of the image that is most relevant for the current decision. Unfortunately, the different methods may disagree and it can be hard to quantify how representative and faithful the explanation really is. We observe however that several of these methods can be seen as edge cases of a single, more general procedure based on finding a particular path through the classifier's domain. This offers additional geometric interpretation to the existing methods. We demonstrate furthermore that ablation paths can be directly used as a technique of its own right. This is able to compete with literature methods on existing benchmarks, while giving more fine-grained information and better opportunities for validation of the explanations' faithfulness.

arXiv:2301.05253

A Scalable Technique for Weak-Supervised Learning with Domain Constraints

We propose a novel scalable end-to-end pipeline that uses symbolic domain knowledge as constraints for learning a neural network for classifying unlabeled data in a weak-supervised manner. Our approach is particularly well-suited for settings where the data consists of distinct groups (classes) that lends itself to clustering-friendly representation learning and the domain constraints can be reformulated for use of efficient mathematical optimization techniques by considering multiple training examples at once. We evaluate our approach on a variant of the MNIST image classification problem where a training example consists of image sequences and the sum of the numbers represented by the sequences, and show that our approach scales significantly better than previous approaches that rely on computing all constraint satisfying combinations for each training example.

arXiv:2304.10582

Invariant Scattering Transform for Medical Imaging

Over the years, the Invariant Scattering Transform (IST) technique has become popular for medical image analysis, including using wavelet transform computation using Convolutional Neural Networks (CNN) to capture patterns' scale and orientation in the input signal. IST aims to be invariant to transformations that are common in medical images, such as translation, rotation, scaling, and deformation, used to improve the performance in medical imaging applications such as segmentation, classification, and registration, which can be integrated into machine learning algorithms for disease detection, diagnosis, and treatment planning. Additionally, combining IST with deep learning approaches has the potential to leverage their strengths and enhance medical image analysis outcomes. This study provides an overview of IST in medical imaging by considering the types of IST, their application, limitations, and potential scopes for future researchers and practitioners.