The Transformer-based method has demonstrated remarkable performance for image super-resolution in comparison to the method based on the convolutional neural networks (CNNs). However, using the self-attention mechanism like SwinIR (Image Restoration Using Swin Transformer) to extract feature information from images needs a significant amount of computational resources, which limits its application on low computing power platforms. To improve the model feature reuse, this research work proposes the Interval Dense Connection Strategy, which connects different blocks according to the newly designed algorithm. We apply this strategy to SwinIR and present a new model, which named SwinOIR (Object Image Restoration Using Swin Transformer). For image super-resolution, an ablation study is conducted to demonstrate the positive effect of the Interval Dense Connection Strategy on the model performance. Furthermore, we evaluate our model on various popular benchmark datasets, and compare it with other state-of-the-art (SOTA) lightweight models. For example, SwinOIR obtains a PSNR of 26.62 dB for x4 upscaling image super-resolution on Urban100 dataset, which is 0.15 dB higher than the SOTA model SwinIR. For real-life application, this work applies the lastest version of You Only Look Once (YOLOv8) model and the proposed model to perform object detection and real-life image super-resolution on low-quality images. This implementation code is publicly available at https://github.com/Rubbbbbbbbby/SwinOIR.
Medical report generation (MRG) is essential for computer-aided diagnosis and medication guidance, which can relieve the heavy burden of radiologists by automatically generating the corresponding medical reports according to the given radiology image. However, due to the spurious correlations within image-text data induced by visual and linguistic biases, it is challenging to generate accurate reports reliably describing lesion areas. Moreover, the cross-modal confounders are usually unobservable and challenging to be eliminated explicitly. In this paper, we aim to mitigate the cross-modal data bias for MRG from a new perspective, i.e., cross-modal causal intervention, and propose a novel Visual-Linguistic Causal Intervention (VLCI) framework for MRG, which consists of a visual deconfounding module (VDM) and a linguistic deconfounding module (LDM), to implicitly mitigate the visual-linguistic confounders by causal front-door intervention. Specifically, due to the absence of a generalized semantic extractor, the VDM explores and disentangles the visual confounders from the patch-based local and global features without expensive fine-grained annotations. Simultaneously, due to the lack of knowledge encompassing the entire field of medicine, the LDM eliminates the linguistic confounders caused by salient visual features and high-frequency context without constructing a terminology database. Extensive experiments on IU-Xray and MIMIC-CXR datasets show that our VLCI significantly outperforms the state-of-the-art MRG methods. The code and models are available at https://github.com/WissingChen/VLCI.
It is widely recognized that deep learning models lack robustness to adversarial examples. An intriguing property of adversarial examples is that they can transfer across different models, which enables black-box attacks without any knowledge of the victim model. An effective strategy to improve the transferability is attacking an ensemble of models. However, previous works simply average the outputs of different models, lacking an in-depth analysis on how and why model ensemble methods can strongly improve the transferability. In this paper, we rethink the ensemble in adversarial attacks and define the common weakness of model ensemble with two properties: 1) the flatness of loss landscape; and 2) the closeness to the local optimum of each model. We empirically and theoretically show that both properties are strongly correlated with the transferability and propose a Common Weakness Attack (CWA) to generate more transferable adversarial examples by promoting these two properties. Experimental results on both image classification and object detection tasks validate the effectiveness of our approach to improving the adversarial transferability, especially when attacking adversarially trained models. We also successfully apply our method to attack a black-box large vision-language model -- Google's Bard, showing the practical effectiveness. Code is available at \url{https://github.com/huanranchen/AdversarialAttacks}.
Scale variation across object instances remains a key challenge in object detection task. Despite the remarkable progress made by modern detection models, this challenge is particularly evident in the semi-supervised case. While existing semi-supervised object detection methods rely on strict conditions to filter high-quality pseudo labels from network predictions, we observe that objects with extreme scale tend to have low confidence, resulting in a lack of positive supervision for these objects. In this paper, we propose a novel framework that addresses the scale variation problem by introducing a mixed scale teacher to improve pseudo label generation and scale-invariant learning. Additionally, we propose mining pseudo labels using score promotion of predictions across scales, which benefits from better predictions from mixed scale features. Our extensive experiments on MS COCO and PASCAL VOC benchmarks under various semi-supervised settings demonstrate that our method achieves new state-of-the-art performance. The code and models are available at \url{https://github.com/lliuz/MixTeacher}.
Recent research on remote sensing object detection has largely focused on improving the representation of oriented bounding boxes but has overlooked the unique prior knowledge presented in remote sensing scenarios. Such prior knowledge can be useful because tiny remote sensing objects may be mistakenly detected without referencing a sufficiently long-range context, and the long-range context required by different types of objects can vary. In this paper, we take these priors into account and propose the Large Selective Kernel Network (LSKNet). LSKNet can dynamically adjust its large spatial receptive field to better model the ranging context of various objects in remote sensing scenarios. To the best of our knowledge, this is the first time that large and selective kernel mechanisms have been explored in the field of remote sensing object detection. Without bells and whistles, LSKNet sets new state-of-the-art scores on standard benchmarks, i.e., HRSC2016 (98.46\% mAP), DOTA-v1.0 (81.85\% mAP) and FAIR1M-v1.0 (47.87\% mAP). Based on a similar technique, we rank 2nd place in 2022 the Greater Bay Area International Algorithm Competition. Code is available at https://github.com/zcablii/Large-Selective-Kernel-Network.
Object detection has been used in a wide range of industries. For example, in autonomous driving, the task of object detection is to accurately and efficiently identify and locate a large number of predefined classes of object instances (vehicles, pedestrians, traffic signs, etc.) from videos of roads. In robotics, the industry robot needs to recognize specific machine elements. In the security field, the camera should accurately recognize each face of people. With the wide application of deep learning, the accuracy and efficiency of object detection have been greatly improved, but object detection based on deep learning still faces challenges. Different applications of object detection have different requirements, including highly accurate detection, multi-category object detection, real-time detection, robustness to occlusions, etc. To address the above challenges, based on extensive literature research, this paper analyzes methods for improving and optimizing mainstream object detection algorithms from the perspective of evolution of one-stage and two-stage object detection algorithms. Furthermore, this article proposes methods for improving object detection accuracy from the perspective of changing receptive fields. The new model is based on the original YOLOv5 (You Look Only Once) with some modifications. The structure of the head part of YOLOv5 is modified by adding asymmetrical pooling layers. As a result, the accuracy of the algorithm is improved while ensuring the speed. The performances of the new model in this article are compared with original YOLOv5 model and analyzed from several parameters. And the evaluation of the new model is presented in four situations. Moreover, the summary and outlooks are made on the problems to be solved and the research directions in the future.
Multiple robots could perceive a scene (e.g., detect objects) collaboratively better than individuals, although easily suffer from adversarial attacks when using deep learning. This could be addressed by the adversarial defense, but its training requires the often-unknown attacking mechanism. Differently, we propose ROBOSAC, a novel sampling-based defense strategy generalizable to unseen attackers. Our key idea is that collaborative perception should lead to consensus rather than dissensus in results compared to individual perception. This leads to our hypothesize-and-verify framework: perception results with and without collaboration from a random subset of teammates are compared until reaching a consensus. In such a framework, more teammates in the sampled subset often entail better perception performance but require longer sampling time to reject potential attackers. Thus, we derive how many sampling trials are needed to ensure the desired size of an attacker-free subset, or equivalently, the maximum size of such a subset that we can successfully sample within a given number of trials. We validate our method on the task of collaborative 3D object detection in autonomous driving scenarios.
Deep learning has become the dominating approach for object detection. To achieve accurate fine-grained detection, one needs to employ a large enough model and a vast amount of data annotations. In this paper, we propose a commonsense knowledge inference module (CKIM) which leverages commonsense knowledge to assist a lightweight deep neural network base coarse-grained object detector to achieve accurate fine-grained detection. Specifically, we focus on a scenario where a single image contains objects of similar categories but varying sizes, and we establish a size-related commonsense knowledge inference module (CKIM) that maps the coarse-grained labels produced by the DL detector to size-related fine-grained labels. Considering that rule-based systems are one of the popular methods of knowledge representation and reasoning, our experiments explored two types of rule-based CKIMs, implemented using crisp-rule and fuzzy-rule approaches, respectively. Experimental results demonstrate that compared with baseline methods, our approach achieves accurate fine-grained detection with a reduced amount of annotated data and smaller model size. Our code is available at: https://github.com/ZJLAB-AMMI/CKIM.
We propose Dataset Reinforcement, a strategy to improve a dataset once such that the accuracy of any model architecture trained on the reinforced dataset is improved at no additional training cost for users. We propose a Dataset Reinforcement strategy based on data augmentation and knowledge distillation. Our generic strategy is designed based on extensive analysis across CNN- and transformer-based models and performing large-scale study of distillation with state-of-the-art models with various data augmentations. We create a reinforced version of the ImageNet training dataset, called ImageNet+, as well as reinforced datasets CIFAR-100+, Flowers-102+, and Food-101+. Models trained with ImageNet+ are more accurate, robust, and calibrated, and transfer well to downstream tasks (e.g., segmentation and detection). As an example, the accuracy of ResNet-50 improves by 1.7% on the ImageNet validation set, 3.5% on ImageNetV2, and 10.0% on ImageNet-R. Expected Calibration Error (ECE) on the ImageNet validation set is also reduced by 9.9%. Using this backbone with Mask-RCNN for object detection on MS-COCO, the mean average precision improves by 0.8%. We reach similar gains for MobileNets, ViTs, and Swin-Transformers. For MobileNetV3 and Swin-Tiny, we observe significant improvements on ImageNet-R/A/C of up to 20% improved robustness. Models pretrained on ImageNet+ and fine-tuned on CIFAR-100+, Flowers-102+, and Food-101+, reach up to 3.4% improved accuracy. The code, datasets, and pretrained models are available at https://github.com/apple/ml-dr.
Video object detection needs to solve feature degradation situations that rarely happen in the image domain. One solution is to use the temporal information and fuse the features from the neighboring frames. With Transformerbased object detectors getting a better performance on the image domain tasks, recent works began to extend those methods to video object detection. However, those existing Transformer-based video object detectors still follow the same pipeline as those used for classical object detectors, like enhancing the object feature representations by aggregation. In this work, we take a different perspective on video object detection. In detail, we improve the qualities of queries for the Transformer-based models by aggregation. To achieve this goal, we first propose a vanilla query aggregation module that weighted averages the queries according to the features of the neighboring frames. Then, we extend the vanilla module to a more practical version, which generates and aggregates queries according to the features of the input frames. Extensive experimental results validate the effectiveness of our proposed methods: On the challenging ImageNet VID benchmark, when integrated with our proposed modules, the current state-of-the-art Transformer-based object detectors can be improved by more than 2.4% on mAP and 4.2% on AP50.
Point cloud sequences are commonly used to accurately detect 3D objects in applications such as autonomous driving. Current top-performing multi-frame detectors mostly follow a Detect-and-Fuse framework, which extracts features from each frame of the sequence and fuses them to detect the objects in the current frame. However, this inevitably leads to redundant computation since adjacent frames are highly correlated. In this paper, we propose an efficient Motion-guided Sequential Fusion (MSF) method, which exploits the continuity of object motion to mine useful sequential contexts for object detection in the current frame. We first generate 3D proposals on the current frame and propagate them to preceding frames based on the estimated velocities. The points-of-interest are then pooled from the sequence and encoded as proposal features. A novel Bidirectional Feature Aggregation (BiFA) module is further proposed to facilitate the interactions of proposal features across frames. Besides, we optimize the point cloud pooling by a voxel-based sampling technique so that millions of points can be processed in several milliseconds. The proposed MSF method achieves not only better efficiency than other multi-frame detectors but also leading accuracy, with 83.12% and 78.30% mAP on the LEVEL1 and LEVEL2 test sets of Waymo Open Dataset, respectively. Codes can be found at \url{https://github.com/skyhehe123/MSF}.
In this paper, we study teacher-student learning from the perspective of data initialization and propose a novel algorithm called Active Teacher(Source code are available at: \url{https://github.com/HunterJ-Lin/ActiveTeacher}) for semi-supervised object detection (SSOD). Active Teacher extends the teacher-student framework to an iterative version, where the label set is partially initialized and gradually augmented by evaluating three key factors of unlabeled examples, including difficulty, information and diversity. With this design, Active Teacher can maximize the effect of limited label information while improving the quality of pseudo-labels. To validate our approach, we conduct extensive experiments on the MS-COCO benchmark and compare Active Teacher with a set of recently proposed SSOD methods. The experimental results not only validate the superior performance gain of Active Teacher over the compared methods, but also show that it enables the baseline network, ie, Faster-RCNN, to achieve 100% supervised performance with much less label expenditure, ie 40% labeled examples on MS-COCO. More importantly, we believe that the experimental analyses in this paper can provide useful empirical knowledge for data annotation in practical applications.
BEV perception is of great importance in the field of autonomous driving, serving as the cornerstone of planning, controlling, and motion prediction. The quality of the BEV feature highly affects the performance of BEV perception. However, taking the noises in camera parameters and LiDAR scans into consideration, we usually obtain BEV representation with harmful noises. Diffusion models naturally have the ability to denoise noisy samples to the ideal data, which motivates us to utilize the diffusion model to get a better BEV representation. In this work, we propose an end-to-end framework, named DiffBEV, to exploit the potential of diffusion model to generate a more comprehensive BEV representation. To the best of our knowledge, we are the first to apply diffusion model to BEV perception. In practice, we design three types of conditions to guide the training of the diffusion model which denoises the coarse samples and refines the semantic feature in a progressive way. What's more, a cross-attention module is leveraged to fuse the context of BEV feature and the semantic content of conditional diffusion model. DiffBEV achieves a 25.9% mIoU on the nuScenes dataset, which is 6.2% higher than the best-performing existing approach. Quantitative and qualitative results on multiple benchmarks demonstrate the effectiveness of DiffBEV in BEV semantic segmentation and 3D object detection tasks. The code will be available soon.
Deep supervision, which involves extra supervisions to the intermediate features of a neural network, was widely used in image classification in the early deep learning era since it significantly reduces the training difficulty and eases the optimization like avoiding gradient vanish over the vanilla training. Nevertheless, with the emergence of normalization techniques and residual connection, deep supervision in image classification was gradually phased out. In this paper, we revisit deep supervision for masked image modeling (MIM) that pre-trains a Vision Transformer (ViT) via a mask-and-predict scheme. Experimentally, we find that deep supervision drives the shallower layers to learn more meaningful representations, accelerates model convergence, and expands attention diversities. Our approach, called DeepMIM, significantly boosts the representation capability of each layer. In addition, DeepMIM is compatible with many MIM models across a range of reconstruction targets. For instance, using ViT-B, DeepMIM on MAE achieves 84.2 top-1 accuracy on ImageNet, outperforming MAE by +0.6. By combining DeepMIM with a stronger tokenizer CLIP, our model achieves state-of-the-art performance on various downstream tasks, including image classification (85.6 top-1 accuracy on ImageNet-1K, outperforming MAE-CLIP by +0.8), object detection (52.8 APbox on COCO) and semantic segmentation (53.1 mIoU on ADE20K). Code and models are available at https://github.com/OliverRensu/DeepMIM.
As the core building block of vision transformers, attention is a powerful tool to capture long-range dependency. However, such power comes at a cost: it incurs a huge computation burden and heavy memory footprint as pairwise token interaction across all spatial locations is computed. A series of works attempt to alleviate this problem by introducing handcrafted and content-agnostic sparsity into attention, such as restricting the attention operation to be inside local windows, axial stripes, or dilated windows. In contrast to these approaches, we propose a novel dynamic sparse attention via bi-level routing to enable a more flexible allocation of computations with content awareness. Specifically, for a query, irrelevant key-value pairs are first filtered out at a coarse region level, and then fine-grained token-to-token attention is applied in the union of remaining candidate regions (\ie, routed regions). We provide a simple yet effective implementation of the proposed bi-level routing attention, which utilizes the sparsity to save both computation and memory while involving only GPU-friendly dense matrix multiplications. Built with the proposed bi-level routing attention, a new general vision transformer, named BiFormer, is then presented. As BiFormer attends to a small subset of relevant tokens in a \textbf{query adaptive} manner without distraction from other irrelevant ones, it enjoys both good performance and high computational efficiency, especially in dense prediction tasks. Empirical results across several computer vision tasks such as image classification, object detection, and semantic segmentation verify the effectiveness of our design. Code is available at \url{https://github.com/rayleizhu/BiFormer}.
In this work, we investigate the problem of sketch-based object localization on natural images, where given a crude hand-drawn sketch of an object, the goal is to localize all the instances of the same object on the target image. This problem proves difficult due to the abstract nature of hand-drawn sketches, variations in the style and quality of sketches, and the large domain gap existing between the sketches and the natural images. To mitigate these challenges, existing works proposed attention-based frameworks to incorporate query information into the image features. However, in these works, the query features are incorporated after the image features have already been independently learned, leading to inadequate alignment. In contrast, we propose a sketch-guided vision transformer encoder that uses cross-attention after each block of the transformer-based image encoder to learn query-conditioned image features leading to stronger alignment with the query sketch. Further, at the output of the decoder, the object and the sketch features are refined to bring the representation of relevant objects closer to the sketch query and thereby improve the localization. The proposed model also generalizes to the object categories not seen during training, as the target image features learned by our method are query-aware. Our localization framework can also utilize multiple sketch queries via a trainable novel sketch fusion strategy. The model is evaluated on the images from the public object detection benchmark, namely MS-COCO, using the sketch queries from QuickDraw! and Sketchy datasets. Compared with existing localization methods, the proposed approach gives a $6.6\%$ and $8.0\%$ improvement in mAP for seen objects using sketch queries from QuickDraw! and Sketchy datasets, respectively, and a $12.2\%$ improvement in AP@50 for large objects that are `unseen' during training.
Monocular 3D object detection has become a mainstream approach in automatic driving for its easy application. A prominent advantage is that it does not need LiDAR point clouds during the inference. However, most current methods still rely on 3D point cloud data for labeling the ground truths used in the training phase. This inconsistency between the training and inference makes it hard to utilize the large-scale feedback data and increases the data collection expenses. To bridge this gap, we propose a new weakly supervised monocular 3D objection detection method, which can train the model with only 2D labels marked on images. To be specific, we explore three types of consistency in this task, i.e. the projection, multi-view and direction consistency, and design a weakly-supervised architecture based on these consistencies. Moreover, we propose a new 2D direction labeling method in this task to guide the model for accurate rotation direction prediction. Experiments show that our weakly-supervised method achieves comparable performance with some fully supervised methods. When used as a pre-training method, our model can significantly outperform the corresponding fully-supervised baseline with only 1/3 3D labels. https://github.com/weakmono3d/weakmono3d
The quadratic computational complexity to the number of tokens limits the practical applications of Vision Transformers (ViTs). Several works propose to prune redundant tokens to achieve efficient ViTs. However, these methods generally suffer from (i) dramatic accuracy drops, (ii) application difficulty in the local vision transformer, and (iii) non-general-purpose networks for downstream tasks. In this work, we propose a novel Semantic Token ViT (STViT), for efficient global and local vision transformers, which can also be revised to serve as backbone for downstream tasks. The semantic tokens represent cluster centers, and they are initialized by pooling image tokens in space and recovered by attention, which can adaptively represent global or local semantic information. Due to the cluster properties, a few semantic tokens can attain the same effect as vast image tokens, for both global and local vision transformers. For instance, only 16 semantic tokens on DeiT-(Tiny,Small,Base) can achieve the same accuracy with more than 100% inference speed improvement and nearly 60% FLOPs reduction; on Swin-(Tiny,Small,Base), we can employ 16 semantic tokens in each window to further speed it up by around 20% with slight accuracy increase. Besides great success in image classification, we also extend our method to video recognition. In addition, we design a STViT-R(ecover) network to restore the detailed spatial information based on the STViT, making it work for downstream tasks, which is powerless for previous token sparsification methods. Experiments demonstrate that our method can achieve competitive results compared to the original networks in object detection and instance segmentation, with over 30% FLOPs reduction for backbone. Code is available at http://github.com/changsn/STViT-R
While most recent autonomous driving system focuses on developing perception methods on ego-vehicle sensors, people tend to overlook an alternative approach to leverage intelligent roadside cameras to extend the perception ability beyond the visual range. We discover that the state-of-the-art vision-centric bird's eye view detection methods have inferior performances on roadside cameras. This is because these methods mainly focus on recovering the depth regarding the camera center, where the depth difference between the car and the ground quickly shrinks while the distance increases. In this paper, we propose a simple yet effective approach, dubbed BEVHeight, to address this issue. In essence, instead of predicting the pixel-wise depth, we regress the height to the ground to achieve a distance-agnostic formulation to ease the optimization process of camera-only perception methods. On popular 3D detection benchmarks of roadside cameras, our method surpasses all previous vision-centric methods by a significant margin. The code is available at {\url{https://github.com/ADLab-AutoDrive/BEVHeight}}.
Self-supervised pre-training and transformer-based networks have significantly improved the performance of object detection. However, most of the current self-supervised object detection methods are built on convolutional-based architectures. We believe that the transformers' sequence characteristics should be considered when designing a transformer-based self-supervised method for the object detection task. To this end, we propose SeqCo-DETR, a novel Sequence Consistency-based self-supervised method for object DEtection with TRansformers. SeqCo-DETR defines a simple but effective pretext by minimizes the discrepancy of the output sequences of transformers with different image views as input and leverages bipartite matching to find the most relevant sequence pairs to improve the sequence-level self-supervised representation learning performance. Furthermore, we provide a mask-based augmentation strategy incorporated with the sequence consistency strategy to extract more representative contextual information about the object for the object detection task. Our method achieves state-of-the-art results on MS COCO (45.8 AP) and PASCAL VOC (64.1 AP), demonstrating the effectiveness of our approach.