PyTorchKR Papers

최신 논문

총 10,167편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2311.05494

Object-centric Cross-modal Feature Distillation for Event-based Object Detection

Event cameras are gaining popularity due to their unique properties, such as their low latency and high dynamic range. One task where these benefits can be crucial is real-time object detection. However, RGB detectors still outperform event-based detectors due to the sparsity of the event data and missing visual details. In this paper, we develop a novel knowledge distillation approach to shrink the performance gap between these two modalities. To this end, we propose a cross-modality object detection distillation method that by design can focus on regions where the knowledge distillation works best. We achieve this by using an object-centric slot attention mechanism that can iteratively decouple features maps into object-centric features and corresponding pixel-features used for distillation. We evaluate our novel distillation approach on a synthetic and a real event dataset with aligned grayscale images as a teacher modality. We show that object-centric distillation allows to significantly improve the performance of the event-based student object detector, nearly halving the performance gap with respect to the teacher.

arXiv:2311.05197

Deep learning in computed tomography pulmonary angiography imaging: a dual-pronged approach for pulmonary embolism detection

The increasing reliance on Computed Tomography Pulmonary Angiography (CTPA) for Pulmonary Embolism (PE) diagnosis presents challenges and a pressing need for improved diagnostic solutions. The primary objective of this study is to leverage deep learning techniques to enhance the Computer Assisted Diagnosis (CAD) of PE. With this aim, we propose a classifier-guided detection approach that effectively leverages the classifier's probabilistic inference to direct the detection predictions, marking a novel contribution in the domain of automated PE diagnosis. Our classification system includes an Attention-Guided Convolutional Neural Network (AG-CNN) that uses local context by employing an attention mechanism. This approach emulates a human expert's attention by looking at both global appearances and local lesion regions before making a decision. The classifier demonstrates robust performance on the FUMPE dataset, achieving an AUROC of 0.927, sensitivity of 0.862, specificity of 0.879, and an F1-score of 0.805 with the Inception-v3 backbone architecture. Moreover, AG-CNN outperforms the baseline DenseNet-121 model, achieving an 8.1% AUROC gain. While previous research has mostly focused on finding PE in the main arteries, our use of cutting-edge object detection models and ensembling techniques greatly improves the accuracy of detecting small embolisms in the peripheral arteries. Finally, our proposed classifier-guided detection approach further refines the detection metrics, contributing new state-of-the-art to the community: mAP$_{50}$, sensitivity, and F1-score of 0.846, 0.901, and 0.779, respectively, outperforming the former benchmark with a significant 3.7% improvement in mAP$_{50}$. Our research aims to elevate PE patient care by integrating AI solutions into clinical workflows, highlighting the potential of human-AI collaboration in medical diagnostics.

arXiv:2311.04558

Free-Space Optical Spiking Neural Network

Neuromorphic engineering has emerged as a promising avenue for developing brain-inspired computational systems. However, conventional electronic AI-based processors often encounter challenges related to processing speed and thermal dissipation. As an alternative, optical implementations of such processors have been proposed, capitalizing on the intrinsic information-processing capabilities of light. Within the realm of optical neuromorphic engineering, various optical neural networks (ONNs) have been explored. Among these, Spiking Neural Networks (SNNs) have exhibited notable success in emulating the computational principles of the human brain. Nevertheless, the integration of optical SNN processors has presented formidable obstacles, mainly when dealing with the computational demands of large datasets. In response to these challenges, we introduce a pioneering concept: the Free-space Optical deep Spiking Convolutional Neural Network (OSCNN). This novel approach draws inspiration from computational models of the human eye. We have meticulously designed various optical components within the OSCNN to tackle object detection tasks across prominent benchmark datasets, including MNIST, ETH 80, and Caltech. Our results demonstrate promising performance with minimal latency and power consumption compared to their electronic ONN counterparts. Additionally, we conducted several pertinent simulations, such as optical intensity to-latency conversion and synchronization. Of particular significance is the evaluation of the feature extraction layer, employing a Gabor filter bank, which stands to impact the practical deployment of diverse ONN architectures significantly.

arXiv:2311.04815

Domain Adaptive Object Detection via Balancing Between Self-Training and Adversarial Learning

Deep learning based object detectors struggle generalizing to a new target domain bearing significant variations in object and background. Most current methods align domains by using image or instance-level adversarial feature alignment. This often suffers due to unwanted background and lacks class-specific alignment. A straightforward approach to promote class-level alignment is to use high confidence predictions on unlabeled domain as pseudo-labels. These predictions are often noisy since model is poorly calibrated under domain shift. In this paper, we propose to leverage model's predictive uncertainty to strike the right balance between adversarial feature alignment and class-level alignment. We develop a technique to quantify predictive uncertainty on class assignments and bounding-box predictions. Model predictions with low uncertainty are used to generate pseudo-labels for self-training, whereas the ones with higher uncertainty are used to generate tiles for adversarial feature alignment. This synergy between tiling around uncertain object regions and generating pseudo-labels from highly certain object regions allows capturing both image and instance-level context during the model adaptation. We report thorough ablation study to reveal the impact of different components in our approach. Results on five diverse and challenging adaptation scenarios show that our approach outperforms existing state-of-the-art methods with noticeable margins.

arXiv:2311.04509

Learning Discriminative Features for Crowd Counting

Crowd counting models in highly congested areas confront two main challenges: weak localization ability and difficulty in differentiating between foreground and background, leading to inaccurate estimations. The reason is that objects in highly congested areas are normally small and high level features extracted by convolutional neural networks are less discriminative to represent small objects. To address these problems, we propose a learning discriminative features framework for crowd counting, which is composed of a masked feature prediction module (MPM) and a supervised pixel-level contrastive learning module (CLM). The MPM randomly masks feature vectors in the feature map and then reconstructs them, allowing the model to learn about what is present in the masked regions and improving the model's ability to localize objects in high density regions. The CLM pulls targets close to each other and pushes them far away from background in the feature space, enabling the model to discriminate foreground objects from background. Additionally, the proposed modules can be beneficial in various computer vision tasks, such as crowd counting and object detection, where dense scenes or cluttered environments pose challenges to accurate localization. The proposed two modules are plug-and-play, incorporating the proposed modules into existing models can potentially boost their performance in these scenarios.

arXiv:2311.04391

3DiffTection: 3D Object Detection with Geometry-Aware Diffusion Features

We present 3DiffTection, a state-of-the-art method for 3D object detection from single images, leveraging features from a 3D-aware diffusion model. Annotating large-scale image data for 3D detection is resource-intensive and time-consuming. Recently, pretrained large image diffusion models have become prominent as effective feature extractors for 2D perception tasks. However, these features are initially trained on paired text and image data, which are not optimized for 3D tasks, and often exhibit a domain gap when applied to the target data. Our approach bridges these gaps through two specialized tuning strategies: geometric and semantic. For geometric tuning, we fine-tune a diffusion model to perform novel view synthesis conditioned on a single image, by introducing a novel epipolar warp operator. This task meets two essential criteria: the necessity for 3D awareness and reliance solely on posed image data, which are readily available (e.g., from videos) and does not require manual annotation. For semantic refinement, we further train the model on target data with detection supervision. Both tuning phases employ ControlNet to preserve the integrity of the original feature capabilities. In the final step, we harness these enhanced capabilities to conduct a test-time prediction ensemble across multiple virtual viewpoints. Through our methodology, we obtain 3D-aware features that are tailored for 3D detection and excel in identifying cross-view point correspondences. Consequently, our model emerges as a powerful 3D detector, substantially surpassing previous benchmarks, e.g., Cube-RCNN, a precedent in single-view 3D detection by 9.43\% in AP3D on the Omni3D-ARkitscene dataset. Furthermore, 3DiffTection showcases robust data efficiency and generalization to cross-domain data.

arXiv:2311.04888

Towards Few-Annotation Learning in Computer Vision: Application to Image Classification and Object Detection tasks

In this thesis, we develop theoretical, algorithmic and experimental contributions for Machine Learning with limited labels, and more specifically for the tasks of Image Classification and Object Detection in Computer Vision. In a first contribution, we are interested in bridging the gap between theory and practice for popular Meta-Learning algorithms used in Few-Shot Classification. We make connections to Multi-Task Representation Learning, which benefits from solid theoretical foundations, to verify the best conditions for a more efficient meta-learning. Then, to leverage unlabeled data when training object detectors based on the Transformer architecture, we propose both an unsupervised pretraining and a semi-supervised learning method in two other separate contributions. For pretraining, we improve Contrastive Learning for object detectors by introducing the localization information. Finally, our semi-supervised method is the first tailored to transformer-based detectors.

arXiv:2311.04828공식 코드 · Papers with Code

SODAWideNet -- Salient Object Detection with an Attention augmented Wide Encoder Decoder network without ImageNet pre-training

Developing a new Salient Object Detection (SOD) model involves selecting an ImageNet pre-trained backbone and creating novel feature refinement modules to use backbone features. However, adding new components to a pre-trained backbone needs retraining the whole network on the ImageNet dataset, which requires significant time. Hence, we explore developing a neural network from scratch directly trained on SOD without ImageNet pre-training. Such a formulation offers full autonomy to design task-specific components. To that end, we propose SODAWideNet, an encoder-decoder-style network for Salient Object Detection. We deviate from the commonly practiced paradigm of narrow and deep convolutional models to a wide and shallow architecture, resulting in a parameter-efficient deep neural network. To achieve a shallower network, we increase the receptive field from the beginning of the network using a combination of dilated convolutions and self-attention. Therefore, we propose Multi Receptive Field Feature Aggregation Module (MRFFAM) that efficiently obtains discriminative features from farther regions at higher resolutions using dilated convolutions. Next, we propose Multi-Scale Attention (MSA), which creates a feature pyramid and efficiently computes attention across multiple resolutions to extract global features from larger feature maps. Finally, we propose two variants, SODAWideNet-S (3.03M) and SODAWideNet (9.03M), that achieve competitive performance against state-of-the-art models on five datasets.

arXiv:2311.03362

Simulation-based Safety Assurance for an AVP System incorporating Learning-Enabled Components

There have been major developments in Automated Driving (AD) and Driving Assist Systems (ADAS) in recent years. However, their safety assurance, thus methodologies for testing, verification and validation AD/ADAS safety-critical applications remain as one the main challenges. Inevitably AI also penetrates into AD/ADAS applications, such as object detection. Despite important benefits, adoption of such learned-enabled components and systems in safety-critical scenarios causes that conventional testing approaches (e.g., distance-based testing in automotive) quickly become infeasible. Similarly, safety engineering approaches usually assume model-based components and do not handle learning-enabled ones well. The authors have participated in the public-funded project FOCETA , and developed an Automated Valet Parking (AVP) use case. As the nature of the baseline implementation is imperfect, it offers a space for continuous improvement based on modelling, verification, validation, and monitoring techniques. In this publication, we explain the simulation-based development platform that is designed to verify and validate safety-critical learning-enabled systems in continuous engineering loops.

arXiv:2311.03762

Image change detection with only a few samples

This paper considers image change detection with only a small number of samples, which is a significant problem in terms of a few annotations available. A major impediment of image change detection task is the lack of large annotated datasets covering a wide variety of scenes. Change detection models trained on insufficient datasets have shown poor generalization capability. To address the poor generalization issue, we propose using simple image processing methods for generating synthetic but informative datasets, and design an early fusion network based on object detection which could outperform the siamese neural network. Our key insight is that the synthetic data enables the trained model to have good generalization ability for various scenarios. We compare the model trained on the synthetic data with that on the real-world data captured from a challenging dataset, CDNet, using six different test sets. The results demonstrate that the synthetic data is informative enough to achieve higher generalization ability than the insufficient real-world data. Besides, the experiment shows that utilizing a few (often tens of) samples to fine-tune the model trained on the synthetic data will achieve excellent results.

arXiv:2311.03648공식 코드 · Papers with Code

Instruct Me More! Random Prompting for Visual In-Context Learning

Large-scale models trained on extensive datasets, have emerged as the preferred approach due to their high generalizability across various tasks. In-context learning (ICL), a popular strategy in natural language processing, uses such models for different tasks by providing instructive prompts but without updating model parameters. This idea is now being explored in computer vision, where an input-output image pair (called an in-context pair) is supplied to the model with a query image as a prompt to exemplify the desired output. The efficacy of visual ICL often depends on the quality of the prompts. We thus introduce a method coined Instruct Me More (InMeMo), which augments in-context pairs with a learnable perturbation (prompt), to explore its potential. Our experiments on mainstream tasks reveal that InMeMo surpasses the current state-of-the-art performance. Specifically, compared to the baseline without learnable prompt, InMeMo boosts mIoU scores by 7.35 and 15.13 for foreground segmentation and single object detection tasks, respectively. Our findings suggest that InMeMo offers a versatile and efficient way to enhance the performance of visual ICL with lightweight training. Code is available at https://github.com/Jackieam/InMeMo.

arXiv:2311.03620

FusionViT: Hierarchical 3D Object Detection via LiDAR-Camera Vision Transformer Fusion

For 3D object detection, both camera and lidar have been demonstrated to be useful sensory devices for providing complementary information about the same scenery with data representations in different modalities, e.g., 2D RGB image vs 3D point cloud. An effective representation learning and fusion of such multi-modal sensor data is necessary and critical for better 3D object detection performance. To solve the problem, in this paper, we will introduce a novel vision transformer-based 3D object detection model, namely FusionViT. Different from the existing 3D object detection approaches, FusionViT is a pure-ViT based framework, which adopts a hierarchical architecture by extending the transformer model to embed both images and point clouds for effective representation learning. Such multi-modal data embedding representations will be further fused together via a fusion vision transformer model prior to feeding the learned features to the object detection head for both detection and localization of the 3D objects in the input scenery. To demonstrate the effectiveness of FusionViT, extensive experiments have been done on real-world traffic object detection benchmark datasets KITTI and Waymo Open. Notably, our FusionViT model can achieve state-of-the-art performance and outperforms not only the existing baseline methods that merely rely on camera images or lidar point clouds, but also the latest multi-modal image-point cloud deep fusion approaches.

arXiv:2311.04058

mmFUSION: Multimodal Fusion for 3D Objects Detection

Multi-sensor fusion is essential for accurate 3D object detection in self-driving systems. Camera and LiDAR are the most commonly used sensors, and usually, their fusion happens at the early or late stages of 3D detectors with the help of regions of interest (RoIs). On the other hand, fusion at the intermediate level is more adaptive because it does not need RoIs from modalities but is complex as the features of both modalities are presented from different points of view. In this paper, we propose a new intermediate-level multi-modal fusion (mmFUSION) approach to overcome these challenges. First, the mmFUSION uses separate encoders for each modality to compute features at a desired lower space volume. Second, these features are fused through cross-modality and multi-modality attention mechanisms proposed in mmFUSION. The mmFUSION framework preserves multi-modal information and learns to complement modalities' deficiencies through attention weights. The strong multi-modal features from the mmFUSION framework are fed to a simple 3D detection head for 3D predictions. We evaluate mmFUSION on the KITTI and NuScenes dataset where it performs better than available early, intermediate, late, and even two-stage based fusion schemes. The code with the mmdetection3D project plugin will be publicly available soon.

arXiv:2311.04040공식 코드 · Papers with Code

Data exploitation: multi-task learning of object detection and semantic segmentation on partially annotated data

Multi-task partially annotated data where each data point is annotated for only a single task are potentially helpful for data scarcity if a network can leverage the inter-task relationship. In this paper, we study the joint learning of object detection and semantic segmentation, the two most popular vision problems, from multi-task data with partial annotations. Extensive experiments are performed to evaluate each task performance and explore their complementarity when a multi-task network cannot optimize both tasks simultaneously. We propose employing knowledge distillation to leverage joint-task optimization. The experimental results show favorable results for multi-task learning and knowledge distillation over single-task learning and even full supervision scenario. All code and data splits are available at https://github.com/lhoangan/multas

arXiv:2311.03774

Meta-Adapter: An Online Few-shot Learner for Vision-Language Model

The contrastive vision-language pre-training, known as CLIP, demonstrates remarkable potential in perceiving open-world visual concepts, enabling effective zero-shot image recognition. Nevertheless, few-shot learning methods based on CLIP typically require offline fine-tuning of the parameters on few-shot samples, resulting in longer inference time and the risk of over-fitting in certain domains. To tackle these challenges, we propose the Meta-Adapter, a lightweight residual-style adapter, to refine the CLIP features guided by the few-shot samples in an online manner. With a few training samples, our method can enable effective few-shot learning capabilities and generalize to unseen data or tasks without additional fine-tuning, achieving competitive performance and high efficiency. Without bells and whistles, our approach outperforms the state-of-the-art online few-shot learning method by an average of 3.6\% on eight image classification datasets with higher inference speed. Furthermore, our model is simple and flexible, serving as a plug-and-play module directly applicable to downstream tasks. Without further fine-tuning, Meta-Adapter obtains notable performance improvements in open-vocabulary object detection and segmentation tasks.

arXiv:2311.03742

3DifFusionDet: Diffusion Model for 3D Object Detection with Robust LiDAR-Camera Fusion

Good 3D object detection performance from LiDAR-Camera sensors demands seamless feature alignment and fusion strategies. We propose the 3DifFusionDet framework in this paper, which structures 3D object detection as a denoising diffusion process from noisy 3D boxes to target boxes. In this framework, ground truth boxes diffuse in a random distribution for training, and the model learns to reverse the noising process. During inference, the model gradually refines a set of boxes that were generated at random to the outcomes. Under the feature align strategy, the progressive refinement method could make a significant contribution to robust LiDAR-Camera fusion. The iterative refinement process could also demonstrate great adaptability by applying the framework to various detecting circumstances where varying levels of accuracy and speed are required. Extensive experiments on KITTI, a benchmark for real-world traffic object identification, revealed that 3DifFusionDet is able to perform favorably in comparison to earlier, well-respected detectors.

arXiv:2311.02083

MaRU: A Manga Retrieval and Understanding System Connecting Vision and Language

Manga, a widely celebrated Japanese comic art form, is renowned for its diverse narratives and distinct artistic styles. However, the inherently visual and intricate structure of Manga, which comprises images housing multiple panels, poses significant challenges for content retrieval. To address this, we present MaRU (Manga Retrieval and Understanding), a multi-staged system that connects vision and language to facilitate efficient search of both dialogues and scenes within Manga frames. The architecture of MaRU integrates an object detection model for identifying text and frame bounding boxes, a Vision Encoder-Decoder model for text recognition, a text encoder for embedding text, and a vision-text encoder that merges textual and visual information into a unified embedding space for scene retrieval. Rigorous evaluations reveal that MaRU excels in end-to-end dialogue retrieval and exhibits promising results for scene retrieval.

arXiv:2311.02535

TokenMotion: Motion-Guided Vision Transformer for Video Camouflaged Object Detection Via Learnable Token Selection

The area of Video Camouflaged Object Detection (VCOD) presents unique challenges in the field of computer vision due to texture similarities between target objects and their surroundings, as well as irregular motion patterns caused by both objects and camera movement. In this paper, we introduce TokenMotion (TMNet), which employs a transformer-based model to enhance VCOD by extracting motion-guided features using a learnable token selection. Evaluated on the challenging MoCA-Mask dataset, TMNet achieves state-of-the-art performance in VCOD. It outperforms the existing state-of-the-art method by a 12.8% improvement in weighted F-measure, an 8.4% enhancement in S-measure, and a 10.7% boost in mean IoU. The results demonstrate the benefits of utilizing motion-guided features via learnable token selection within a transformer-based framework to tackle the intricate task of VCOD.

arXiv:2311.02342공식 코드 · Papers with Code

Proposal-Level Unsupervised Domain Adaptation for Open World Unbiased Detector

Open World Object Detection (OWOD) combines open-set object detection with incremental learning capabilities to handle the challenge of the open and dynamic visual world. Existing works assume that a foreground predictor trained on the seen categories can be directly transferred to identify the unseen categories' locations by selecting the top-k most confident foreground predictions. However, the assumption is hardly valid in practice. This is because the predictor is inevitably biased to the known categories, and fails under the shift in the appearance of the unseen categories. In this work, we aim to build an unbiased foreground predictor by re-formulating the task under Unsupervised Domain Adaptation, where the current biased predictor helps form the domains: the seen object locations and confident background locations as the source domain, and the rest ambiguous ones as the target domain. Then, we adopt the simple and effective self-training method to learn a predictor based on the domain-invariant foreground features, hence achieving unbiased prediction robust to the shift in appearance between the seen and unseen categories. Our approach's pipeline can adapt to various detection frameworks and UDA methods, empirically validated by OWOD evaluation, where we achieve state-of-the-art performance.

arXiv:2311.02274공식 코드 · Papers with Code

Patch-based Selection and Refinement for Early Object Detection

Early object detection (OD) is a crucial task for the safety of many dynamic systems. Current OD algorithms have limited success for small objects at a long distance. To improve the accuracy and efficiency of such a task, we propose a novel set of algorithms that divide the image into patches, select patches with objects at various scales, elaborate the details of a small object, and detect it as early as possible. Our approach is built upon a transformer-based network and integrates the diffusion model to improve the detection accuracy. As demonstrated on BDD100K, our algorithms enhance the mAP for small objects from 1.03 to 8.93, and reduce the data volume in computation by more than 77\%. The source code is available at \href{https://github.com/destiny301/dpr}{https://github.com/destiny301/dpr}