PyTorchKR Papers

최신 논문

총 10,566편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2206.06067공식 코드 · Papers with Code

Better Teacher Better Student: Dynamic Prior Knowledge for Knowledge Distillation

Knowledge distillation (KD) has shown very promising capabilities in transferring learning representations from large models (teachers) to small models (students). However, as the capacity gap between students and teachers becomes larger, existing KD methods fail to achieve better results. Our work shows that the `prior knowledge' is vital to KD, especially when applying large teachers. Particularly, we propose the dynamic prior knowledge (DPK), which integrates part of teacher's features as the prior knowledge before the feature distillation. This means that our method also takes the teacher's feature as `input', not just `target'. Besides, we dynamically adjust the ratio of the prior knowledge during the training phase according to the feature gap, thus guiding the student in an appropriate difficulty. To evaluate the proposed method, we conduct extensive experiments on two image classification benchmarks (i.e. CIFAR100 and ImageNet) and an object detection benchmark (i.e. MS COCO. The results demonstrate the superiority of our method in performance under varying settings. Besides, our DPK makes the performance of the student model positively correlated with that of the teacher model, which means that we can further boost the accuracy of students by applying larger teachers. More importantly, DPK provides a fast solution in teacher model selection for any given model.

arXiv:2206.05927공식 코드 · Papers with Code

LinK3D: Linear Keypoints Representation for 3D LiDAR Point Cloud

Feature extraction and matching are the basic parts of many robotic vision tasks, such as 2D or 3D object detection, recognition, and registration. As is known, 2D feature extraction and matching have already achieved great success. Unfortunately, in the field of 3D, the current methods may fail to support the extensive application of 3D LiDAR sensors in robotic vision tasks due to their poor descriptiveness and inefficiency. To address this limitation, we propose a novel 3D feature representation method: Linear Keypoints representation for 3D LiDAR point cloud, called LinK3D. The novelty of LinK3D lies in that it fully considers the characteristics (such as the sparsity and complexity) of LiDAR point clouds and represents the keypoint with its robust neighbor keypoints, which provide strong constraints in the description of the keypoint. The proposed LinK3D has been evaluated on three public datasets, and the experimental results show that our method achieves great matching performance. More importantly, LinK3D also shows excellent real-time performance, faster than the sensor frame rate at 10 Hz of a typical rotating LiDAR sensor. LinK3D only takes an average of 30 milliseconds to extract features from the point cloud collected by a 64-beam LiDAR and takes merely about 20 milliseconds to match two LiDAR scans when executed on a computer with an Intel Core i7 processor. Moreover, our method can be extended to LiDAR odometry task, and shows good scalability. We release the implementation of our method at https://github.com/YungeCui/LinK3D.

arXiv:2206.05836공식 코드 · Papers with Code

GLIPv2: Unifying Localization and Vision-Language Understanding

We present GLIPv2, a grounded VL understanding model, that serves both localization tasks (e.g., object detection, instance segmentation) and Vision-Language (VL) understanding tasks (e.g., VQA, image captioning). GLIPv2 elegantly unifies localization pre-training and Vision-Language Pre-training (VLP) with three pre-training tasks: phrase grounding as a VL reformulation of the detection task, region-word contrastive learning as a novel region-word level contrastive learning task, and the masked language modeling. This unification not only simplifies the previous multi-stage VLP procedure but also achieves mutual benefits between localization and understanding tasks. Experimental results show that a single GLIPv2 model (all model weights are shared) achieves near SoTA performance on various localization and understanding tasks. The model also shows (1) strong zero-shot and few-shot adaption performance on open-vocabulary object detection tasks and (2) superior grounding capability on VL understanding tasks. Code will be released at https://github.com/microsoft/GLIP.

arXiv:2206.05765

A Semantic Consistency Feature Alignment Object Detection Model Based on Mixed-Class Distribution Metrics

Unsupervised domain adaptation is critical in various computer vision tasks, such as object detection, instance segmentation, etc. They attempt to reduce domain bias-induced performance degradation while also promoting model application speed. Previous works in domain adaptation object detection attempt to align image-level and instance-level shifts to eventually minimize the domain discrepancy, but they may align single-class features to mixed-class features in image-level domain adaptation because each image in the object detection task may be more than one class and object. In order to achieve single-class with single-class alignment and mixed-class with mixed-class alignment, we treat the mixed-class of the feature as a new class and propose a mixed-classes $H-divergence$ for object detection to achieve homogenous feature alignment and reduce negative transfer. Then, a Semantic Consistency Feature Alignment Model (SCFAM) based on mixed-classes $H-divergence$ was also presented. To improve single-class and mixed-class semantic information and accomplish semantic separation, the SCFAM model proposes Semantic Prediction Models (SPM) and Semantic Bridging Components (SBC). And the weight of the pix domain discriminator loss is then changed based on the SPM result to reduce sample imbalance. Extensive unsupervised domain adaption experiments on widely used datasets illustrate our proposed approach's robust object detection in domain bias settings.

arXiv:2206.05365

Object Detection, Recognition, Deep Learning, and the Universal Law of Generalization

Object detection and recognition are fundamental functions underlying the success of species. Because the appearance of an object exhibits a large variability, the brain has to group these different stimuli under the same object identity, a process of generalization. Does the process of generalization follow some general principles or is it an ad-hoc "bag-of-tricks"? The Universal Law of Generalization provided evidence that generalization follows similar properties across a variety of species and tasks. Here we test the hypothesis that the internal representations underlying generalization reflect the natural properties of object detection and recognition in our environment rather than the specifics of the system solving these problems. By training a deep-neural-network with images of "clear" and "camouflaged" animals, we found that with a proper choice of category prototypes, the generalization functions are monotone decreasing, similar to the generalization functions of biological systems. Our findings support the hypothesis of the study.

arXiv:2206.06293공식 코드 · Papers with Code

Learning Domain Adaptive Object Detection with Probabilistic Teacher

Self-training for unsupervised domain adaptive object detection is a challenging task, of which the performance depends heavily on the quality of pseudo boxes. Despite the promising results, prior works have largely overlooked the uncertainty of pseudo boxes during self-training. In this paper, we present a simple yet effective framework, termed as Probabilistic Teacher (PT), which aims to capture the uncertainty of unlabeled target data from a gradually evolving teacher and guides the learning of a student in a mutually beneficial manner. Specifically, we propose to leverage the uncertainty-guided consistency training to promote classification adaptation and localization adaptation, rather than filtering pseudo boxes via an elaborate confidence threshold. In addition, we conduct anchor adaptation in parallel with localization adaptation, since anchor can be regarded as a learnable parameter. Together with this framework, we also present a novel Entropy Focal Loss (EFL) to further facilitate the uncertainty-guided self-training. Equipped with EFL, PT outperforms all previous baselines by a large margin and achieve new state-of-the-arts.

arXiv:2206.05102

Saccade Mechanisms for Image Classification, Object Detection and Tracking

We examine how the saccade mechanism from biological vision can be used to make deep neural networks more efficient for classification and object detection problems. Our proposed approach is based on the ideas of attention-driven visual processing and saccades, miniature eye movements influenced by attention. We conduct experiments by analyzing: i) the robustness of different deep neural network (DNN) feature extractors to partially-sensed images for image classification and object detection, and ii) the utility of saccades in masking image patches for image classification and object tracking. Experiments with convolutional nets (ResNet-18) and transformer-based models (ViT, DETR, TransTrack) are conducted on several datasets (CIFAR-10, DAVSOD, MSCOCO, and MOT17). Our experiments show intelligent data reduction via learning to mimic human saccades when used in conjunction with state-of-the-art DNNs for classification, detection, and tracking tasks. We observed minimal drop in performance for the classification and detection tasks while only using about 30\% of the original sensor data. We discuss how the saccade mechanism can inform hardware design via ``in-pixel'' processing.

arXiv:2206.05028

Spatial Cross-Attention Improves Self-Supervised Visual Representation Learning

Unsupervised representation learning methods like SwAV are proved to be effective in learning visual semantics of a target dataset. The main idea behind these methods is that different views of a same image represent the same semantics. In this paper, we further introduce an add-on module to facilitate the injection of the knowledge accounting for spatial cross correlations among the samples. This in turn results in distilling intra-class information including feature level locations and cross similarities between same-class instances. The proposed add-on can be added to existing methods such as the SwAV. We can later remove the add-on module for inference without any modification of the learned weights. Through an extensive set of empirical evaluations, we verify that our method yields an improved performance in detecting the class activation maps, top-1 classification accuracy, and down-stream tasks such as object detection, with different configuration settings.

논문 ID: document/9506549

Pseudo-Label Generation-Evaluation Framework For Cross Domain Weakly Supervised Object Detection

Cross domain weakly supervised object detection (CDWSOD), where we can get access to instance-level annotations in the source domain while only image-level annotations are available in the target domain, adapts object detectors from label-rich to label-poor domains. It usually generates pseudo labels in the target domain and utilize them to finetune the detector pretrained in the source domain. In this paper, we propose a new pseudo-label generation-evaluation framework for CDWSOD task. In particular, an evaluator is introduced for the generated pseudo labels in the target domain and the transferring process involves two players: the detector to generate instance-level pseudo labels and the evaluator to judge the quality of pseudo labels. Only high-quality pseudo labels selected by the evaluator are utilized to finetune the detector. Experiments on three representative datasets demonstrate the effectiveness of our framework in various domains.

arXiv:2206.04042공식 코드 · Papers with Code

Learning Ego 3D Representation as Ray Tracing

A self-driving perception model aims to extract 3D semantic representations from multiple cameras collectively into the bird's-eye-view (BEV) coordinate frame of the ego car in order to ground downstream planner. Existing perception methods often rely on error-prone depth estimation of the whole scene or learning sparse virtual 3D representations without the target geometry structure, both of which remain limited in performance and/or capability. In this paper, we present a novel end-to-end architecture for ego 3D representation learning from an arbitrary number of unconstrained camera views. Inspired by the ray tracing principle, we design a polarized grid of "imaginary eyes" as the learnable ego 3D representation and formulate the learning process with the adaptive attention mechanism in conjunction with the 3D-to-2D projection. Critically, this formulation allows extracting rich 3D representation from 2D images without any depth supervision, and with the built-in geometry structure consistent w.r.t. BEV. Despite its simplicity and versatility, extensive experiments on standard BEV visual tasks (e.g., camera-based 3D object detection and BEV segmentation) show that our model outperforms all state-of-the-art alternatives significantly, with an extra advantage in computational efficiency from multi-task learning.

arXiv:2206.04040공식 코드 · Papers with Code

MobileOne: An Improved One millisecond Mobile Backbone

Efficient neural network backbones for mobile devices are often optimized for metrics such as FLOPs or parameter count. However, these metrics may not correlate well with latency of the network when deployed on a mobile device. Therefore, we perform extensive analysis of different metrics by deploying several mobile-friendly networks on a mobile device. We identify and analyze architectural and optimization bottlenecks in recent efficient neural networks and provide ways to mitigate these bottlenecks. To this end, we design an efficient backbone MobileOne, with variants achieving an inference time under 1 ms on an iPhone12 with 75.9% top-1 accuracy on ImageNet. We show that MobileOne achieves state-of-the-art performance within the efficient architectures while being many times faster on mobile. Our best model obtains similar performance on ImageNet as MobileFormer while being 38x faster. Our model obtains 2.3% better top-1 accuracy on ImageNet than EfficientNet at similar latency. Furthermore, we show that our model generalizes to multiple tasks - image classification, object detection, and semantic segmentation with significant improvements in latency and accuracy as compared to existing efficient architectures when deployed on a mobile device. Code and models are available at https://github.com/apple/ml-mobileone

arXiv:2206.03943공식 코드 · Papers with Code

Robust Environment Perception for Automated Driving: A Unified Learning Pipeline for Visual-Infrared Object Detection

The RGB complementary metal-oxidesemiconductor (CMOS) sensor works within the visible light spectrum. Therefore it is very sensitive to environmental light conditions. On the contrary, a long-wave infrared (LWIR) sensor operating in 8-14 micro meter spectral band, functions independent of visible light. In this paper, we exploit both visual and thermal perception units for robust object detection purposes. After delicate synchronization and (cross-) labeling of the FLIR [1] dataset, this multi-modal perception data passes through a convolutional neural network (CNN) to detect three critical objects on the road, namely pedestrians, bicycles, and cars. After evaluation of RGB and infrared (thermal and infrared are often used interchangeably) sensors separately, various network structures are compared to fuse the data at the feature level effectively. Our RGB-thermal (RGBT) fusion network, which takes advantage of a novel entropy-block attention module (EBAM), outperforms the state-of-the-art network [2] by 10% with 82.9% mAP.

arXiv:2206.03657

Delving into the Pre-training Paradigm of Monocular 3D Object Detection

The labels of monocular 3D object detection (M3OD) are expensive to obtain. Meanwhile, there usually exists numerous unlabeled data in practical applications, and pre-training is an efficient way of exploiting the knowledge in unlabeled data. However, the pre-training paradigm for M3OD is hardly studied. We aim to bridge this gap in this work. To this end, we first draw two observations: (1) The guideline of devising pre-training tasks is imitating the representation of the target task. (2) Combining depth estimation and 2D object detection is a promising M3OD pre-training baseline. Afterwards, following the guideline, we propose several strategies to further improve this baseline, which mainly include target guided semi-dense depth estimation, keypoint-aware 2D object detection, and class-level loss adjustment. Combining all the developed techniques, the obtained pre-training framework produces pre-trained backbones that improve M3OD performance significantly on both the KITTI-3D and nuScenes benchmarks. For example, by applying a DLA34 backbone to a naive center-based M3OD detector, the moderate ${\rm AP}_{3D}70$ score of Car on the KITTI-3D testing set is boosted by 18.71\% and the NDS score on the nuScenes validation set is improved by 40.41\% relatively.

arXiv:2206.03105

Dual Swin-Transformer based Mutual Interactive Network for RGB-D Salient Object Detection

Salient Object Detection is the task of predicting the human attended region in a given scene. Fusing depth information has been proven effective in this task. The main challenge of this problem is how to aggregate the complementary information from RGB modality and depth modality. However, conventional deep models heavily rely on CNN feature extractors, and the long-range contextual dependencies are usually ignored. In this work, we propose Dual Swin-Transformer based Mutual Interactive Network. We adopt Swin-Transformer as the feature extractor for both RGB and depth modality to model the long-range dependencies in visual inputs. Before fusing the two branches of features into one, attention-based modules are applied to enhance features from each modality. We design a self-attention-based cross-modality interaction module and a gated modality attention module to leverage the complementary information between the two modalities. For the saliency decoding, we create different stages enhanced with dense connections and keep a decoding memory while the multi-level encoding features are considered simultaneously. Considering the inaccurate depth map issue, we collect the RGB features of early stages into a skip convolution module to give more guidance from RGB modality to the final saliency prediction. In addition, we add edge supervision to regularize the feature learning process. Comprehensive experiments on five standard RGB-D SOD benchmark datasets over four evaluation metrics demonstrate the superiority of the proposed DTMINet method.

arXiv:2206.02977

DETR++: Taming Your Multi-Scale Detection Transformer

Convolutional Neural Networks (CNN) have dominated the field of detection ever since the success of AlexNet in ImageNet classification [12]. With the sweeping reform of Transformers [27] in natural language processing, Carion et al. [2] introduce the Transformer-based detection method, i.e., DETR. However, due to the quadratic complexity in the self-attention mechanism in the Transformer, DETR is never able to incorporate multi-scale features as performed in existing CNN-based detectors, leading to inferior results in small object detection. To mitigate this issue and further improve performance of DETR, in this work, we investigate different methods to incorporate multi-scale features and find that a Bi-directional Feature Pyramid (BiFPN) works best with DETR in further raising the detection precision. With this discovery, we propose DETR++, a new architecture that improves detection results by 1.9% AP on MS COCO 2017, 11.5% AP on RICO icon detection, and 9.1% AP on RICO layout extraction over existing baselines.

arXiv:2206.02876

SpikiLi: A Spiking Simulation of LiDAR based Real-time Object Detection for Autonomous Driving

Spiking Neural Networks are a recent and new neural network design approach that promises tremendous improvements in power efficiency, computation efficiency, and processing latency. They do so by using asynchronous spike-based data flow, event-based signal generation, processing, and modifying the neuron model to resemble biological neurons closely. While some initial works have shown significant initial evidence of applicability to common deep learning tasks, their applications in complex real-world tasks has been relatively low. In this work, we first illustrate the applicability of spiking neural networks to a complex deep learning task namely Lidar based 3D object detection for automated driving. Secondly, we make a step-by-step demonstration of simulating spiking behavior using a pre-trained convolutional neural network. We closely model essential aspects of spiking neural networks in simulation and achieve equivalent run-time and accuracy on a GPU. When the model is realized on a neuromorphic hardware, we expect to have significantly improved power efficiency.

arXiv:2206.03484

Detection Hub: Unifying Object Detection Datasets via Query Adaptation on Language Embedding

Combining multiple datasets enables performance boost on many computer vision tasks. But similar trend has not been witnessed in object detection when combining multiple datasets due to two inconsistencies among detection datasets: taxonomy difference and domain gap. In this paper, we address these challenges by a new design (named Detection Hub) that is dataset-aware and category-aligned. It not only mitigates the dataset inconsistency but also provides coherent guidance for the detector to learn across multiple datasets. In particular, the dataset-aware design is achieved by learning a dataset embedding that is used to adapt object queries as well as convolutional kernels in detection heads. The categories across datasets are semantically aligned into a unified space by replacing one-hot category representations with word embedding and leveraging the semantic coherence of language embedding. Detection Hub fulfills the benefits of large data on object detection. Experiments demonstrate that joint training on multiple datasets achieves significant performance gains over training on each dataset alone. Detection Hub further achieves SoTA performance on UODB benchmark with wide variety of datasets.

arXiv:2206.03382공식 코드 · Papers with Code

Tutel: Adaptive Mixture-of-Experts at Scale

Sparsely-gated mixture-of-experts (MoE) has been widely adopted to scale deep learning models to trillion-plus parameters with fixed computational cost. The algorithmic performance of MoE relies on its token routing mechanism that forwards each input token to the right sub-models or experts. While token routing dynamically determines the amount of expert workload at runtime, existing systems suffer inefficient computation due to their static execution, namely static parallelism and pipelining, which does not adapt to the dynamic workload. We present Flex, a highly scalable stack design and implementation for MoE with dynamically adaptive parallelism and pipelining. Flex designs an identical layout for distributing MoE model parameters and input data, which can be leveraged by all possible parallelism or pipelining methods without any mathematical inequivalence or tensor migration overhead. This enables adaptive parallelism/pipelining optimization at zero cost during runtime. Based on this key design, Flex also implements various MoE acceleration techniques. Aggregating all techniques, Flex finally delivers huge speedup at any scale -- 4.96x and 5.75x speedup of a single MoE layer over 16 and 2,048 A100 GPUs, respectively, over the previous state-of-the-art. Our evaluation shows that Flex efficiently and effectively runs a real-world MoE-based model named SwinV2-MoE, built upon Swin Transformer V2, a state-of-the-art computer vision architecture. On efficiency, Flex accelerates SwinV2-MoE, achieving up to 1.55x and 2.11x speedup in training and inference over Fairseq, respectively. On effectiveness, the SwinV2-MoE model achieves superior accuracy in both pre-training and down-stream computer vision tasks such as COCO object detection than the counterpart dense model, indicating the readiness of Flex for end-to-end real-world model training and inference.

arXiv:2206.02572

Automated visual inspection of silicon detectors in CMS experiment

In the CMS experiment at CERN, Geneva, a large number of HGCAL sensor modules are fabricated in advanced laboratories around the world. Each sensor module contains about 700 checkpoints for visual inspection thus making it almost impossible to carry out such inspection manually. As artificial intelligence is more and more widely used in manufacturing, traditional detection technologies are gradually being intelligent. In order to more accurately evaluate the checkpoints, we propose to use deep learning-based object detection techniques to detect manufacturing defects in testing large numbers of modules automatically.

arXiv:2206.02777공식 코드 · Papers with Code

Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation

In this paper we present Mask DINO, a unified object detection and segmentation framework. Mask DINO extends DINO (DETR with Improved Denoising Anchor Boxes) by adding a mask prediction branch which supports all image segmentation tasks (instance, panoptic, and semantic). It makes use of the query embeddings from DINO to dot-product a high-resolution pixel embedding map to predict a set of binary masks. Some key components in DINO are extended for segmentation through a shared architecture and training process. Mask DINO is simple, efficient, and scalable, and it can benefit from joint large-scale detection and segmentation datasets. Our experiments show that Mask DINO significantly outperforms all existing specialized segmentation methods, both on a ResNet-50 backbone and a pre-trained model with SwinL backbone. Notably, Mask DINO establishes the best results to date on instance segmentation (54.5 AP on COCO), panoptic segmentation (59.4 PQ on COCO), and semantic segmentation (60.8 mIoU on ADE20K) among models under one billion parameters. Code is available at \url{https://github.com/IDEACVR/MaskDINO}.