Tiny Object Detection is challenging due to small size, low resolution, occlusion, background clutter, lighting conditions and small object-to-image ratio. Further, object detection methodologies often make underlying assumption that both training and testing data remain congruent. However, this presumption often leads to decline in performance when model is applied to out-of-domain(unseen) data. Techniques like synthetic image generation are employed to improve model performance by leveraging variations in input data. Such an approach typically presumes access to 3D-rendered datasets. In contrast, we propose a novel two-stage methodology Synthetic Randomized Image Augmentation (SRIA), carefully devised to enhance generalization capabilities of models encountering 2D datasets, particularly with lower resolution which is more practical in real-world scenarios. The first stage employs a weakly supervised technique to generate pixel-level segmentation masks. Subsequently, the second stage generates a batch-wise synthesis of artificial images, carefully designed with an array of diverse augmentations. The efficacy of proposed technique is illustrated on challenging foreign object debris (FOD) detection. We compare our results with several SOTA models including CenterNet, SSD, YOLOv3, YOLOv4, YOLOv5, and Outer Vit on a publicly available FOD-A dataset. We also construct an out-of-distribution test set encompassing 800 annotated images featuring a corpus of ten common categories. Notably, by harnessing merely 1.81% of objects from source training data and amalgamating with 29 runway background images, we generate 2227 synthetic images. Subsequent model retraining via transfer learning, utilizing enriched dataset generated by domain randomization, demonstrates significant improvement in detection accuracy. We report that detection accuracy improved from an initial 41% to 92% for OOD test set.
Single object tracking (SOT) heavily relies on the representation of the target object as a bounding box. However, due to the potential deformation and rotation experienced by the tracked targets, the genuine bounding box fails to capture the appearance information explicitly and introduces cluttered background. This paper proposes RTrack, a novel object representation baseline tracker that utilizes a set of sample points to get a pseudo bounding box. RTrack automatically arranges these points to define the spatial extents and highlight local areas. Building upon the baseline, we conducted an in-depth exploration of the training potential and introduced a one-to-many leading assignment strategy. It is worth noting that our approach achieves competitive performance to the state-of-the-art trackers on the GOT-10k dataset while reducing training time to just 10% of the previous state-of-the-art (SOTA) trackers' training costs. The substantial reduction in training costs brings single-object tracking (SOT) closer to the object detection (OD) task. Extensive experiments demonstrate that our proposed RTrack achieves SOTA results with faster convergence.
Domain adaptation, which aims to transfer knowledge between domains, has been well studied in many areas such as image classification and object detection. However, for multi-modal tasks, conventional approaches rely on large-scale pre-training. But due to the difficulty of acquiring multi-modal data, large-scale pre-training is often impractical. Therefore, domain adaptation, which can efficiently utilize the knowledge from different datasets (domains), is crucial for multi-modal tasks. In this paper, we focus on the Referring Expression Grounding (REG) task, which is to localize an image region described by a natural language expression. Specifically, we propose a novel approach to effectively transfer multi-modal knowledge through a specially relation-tailored approach for the REG problem. Our approach tackles the multi-modal domain adaptation problem by simultaneously enriching inter-domain relations and transferring relations between domains. Experiments show that our proposed approach significantly improves the transferability of multi-modal domains and enhances adaptation performance in the REG problem.
This paper presents CLUSTERFORMER, a universal vision model that is based on the CLUSTERing paradigm with TransFORMER. It comprises two novel designs: 1. recurrent cross-attention clustering, which reformulates the cross-attention mechanism in Transformer and enables recursive updates of cluster centers to facilitate strong representation learning; and 2. feature dispatching, which uses the updated cluster centers to redistribute image features through similarity-based metrics, resulting in a transparent pipeline. This elegant design streamlines an explainable and transferable workflow, capable of tackling heterogeneous vision tasks (i.e., image classification, object detection, and image segmentation) with varying levels of clustering granularity (i.e., image-, box-, and pixel-level). Empirical results demonstrate that CLUSTERFORMER outperforms various well-known specialized architectures, achieving 83.41% top-1 acc. over ImageNet-1K for image classification, 54.2% and 47.0% mAP over MS COCO for object detection and instance segmentation, 52.4% mIoU over ADE20K for semantic segmentation, and 55.8% PQ over COCO Panoptic for panoptic segmentation. For its efficacy, we hope our work can catalyze a paradigm shift in universal models in computer vision.
With the advancing technology, the hardware gain of computers and the increase in the processing capacity of processors have facilitated the processing of instantaneous and real-time images. Face recognition processes are also studies in the field of image processing. Facial recognition processes are frequently used in security applications and commercial applications. Especially in the last 20 years, the high performances of artificial intelligence (AI) studies have contributed to the spread of these studies in many different fields. Education is one of them. The potential and advantages of using AI in education; can be grouped under three headings: student, teacher, and institution. One of the institutional studies may be the security of educational environments and the contribution of automation to education and training processes. From this point of view, deep learning methods, one of the sub-branches of AI, were used in this study. For object detection from images, a pioneering study has been designed and successfully implemented to keep records of students' entrance to the educational institution and to perform class attendance with images taken from the camera using image processing algorithms. The application of the study to real-life problems will be carried out in a school determined in the 2022-2023 academic year.
The shift towards electrification and autonomous driving in the automotive industry results in more and more automotive wire harnesses being installed in modern automobiles, which stresses the great significance of guaranteeing the quality of automotive wire harness assembly. The mating of connectors is essential in the final assembly of automotive wire harnesses due to the importance of connectors on wire harness connection and signal transmission. However, the current manual operation of mating connectors leads to severe problems regarding assembly quality and ergonomics, where the robotized assembly has been considered, and different vision-based solutions have been proposed to facilitate a better perception of the robot control system on connectors. Nonetheless, there has been a lack of deep learning-based solutions for detecting automotive wire harness connectors in previous literature. This paper presents a deep learning-based connector detection for robotized automotive wire harness assembly. A dataset of twenty automotive wire harness connectors was created to train and evaluate a two-stage and a one-stage object detection model, respectively. The experiment results indicate the effectiveness of deep learning-based connector detection for automotive wire harness assembly but are limited by the design of the exteriors of connectors.
Detecting humans from airborne visual and thermal imagery is a fundamental challenge for Wilderness Search-and-Rescue (WiSAR) teams, who must perform this function accurately in the face of immense pressure. The ability to fuse these two sensor modalities can potentially reduce the cognitive load on human operators and/or improve the effectiveness of computer vision object detection models. However, the fusion task is particularly challenging in the context of WiSAR due to hardware limitations and extreme environmental factors. This work presents Misaligned Image Synthesis and Fusion using Information from Thermal and Visual (MISFIT-V), a novel two-pronged unsupervised deep learning approach that utilizes a Generative Adversarial Network (GAN) and a cross-attention mechanism to capture the most relevant features from each modality. Experimental results show MISFIT-V offers enhanced robustness against misalignment and poor lighting/thermal environmental conditions compared to existing visual-thermal image fusion methods.
Self-attention-based vision transformers (ViTs) have emerged as a highly competitive architecture in computer vision. Unlike convolutional neural networks (CNNs), ViTs are capable of global information sharing. With the development of various structures of ViTs, ViTs are increasingly advantageous for many vision tasks. However, the quadratic complexity of self-attention renders ViTs computationally intensive, and their lack of inductive biases of locality and translation equivariance demands larger model sizes compared to CNNs to effectively learn visual features. In this paper, we propose a light-weight and efficient vision transformer model called DualToken-ViT that leverages the advantages of CNNs and ViTs. DualToken-ViT effectively fuses the token with local information obtained by convolution-based structure and the token with global information obtained by self-attention-based structure to achieve an efficient attention structure. In addition, we use position-aware global tokens throughout all stages to enrich the global information, which further strengthening the effect of DualToken-ViT. Position-aware global tokens also contain the position information of the image, which makes our model better for vision tasks. We conducted extensive experiments on image classification, object detection and semantic segmentation tasks to demonstrate the effectiveness of DualToken-ViT. On the ImageNet-1K dataset, our models of different scales achieve accuracies of 75.4% and 79.4% with only 0.5G and 1.0G FLOPs, respectively, and our model with 1.0G FLOPs outperforms LightViT-T using global tokens by 0.7%.
You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level routing attention, Generalized feature pyramid networks, and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP$_{50}$ compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.
Few-shot object detection aims at detecting novel categories given only a few example images. It is a basic skill for a robot to perform tasks in open environments. Recent methods focus on finetuning strategies, with complicated procedures that prohibit a wider application. In this paper, we introduce DE-ViT, a few-shot object detector without the need for finetuning. DE-ViT's novel architecture is based on a new region-propagation mechanism for localization. The propagated region masks are transformed into bounding boxes through a learnable spatial integral layer. Instead of training prototype classifiers, we propose to use prototypes to project ViT features into a subspace that is robust to overfitting on base classes. We evaluate DE-ViT on few-shot, and one-shot object detection benchmarks with Pascal VOC, COCO, and LVIS. DE-ViT establishes new state-of-the-art results on all benchmarks. Notably, for COCO, DE-ViT surpasses the few-shot SoTA by 15 mAP on 10-shot and 7.2 mAP on 30-shot and one-shot SoTA by 2.8 AP50. For LVIS, DE-ViT outperforms few-shot SoTA by 17 box APr. Further, we evaluate DE-ViT with a real robot by building a pick-and-place system for sorting novel objects based on example images. The videos of our robot demonstrations, the source code and the models of DE-ViT can be found at https://mlzxy.github.io/devit.
Recently, end-to-end object detectors have gained significant attention from the research community due to their outstanding performance. However, DETR typically relies on supervised pretraining of the backbone on ImageNet, which limits the practical application of DETR and the design of the backbone, affecting the model's potential generalization ability. In this paper, we propose a new training method called step-by-step training. Specifically, in the first stage, the one-to-many pre-trained YOLO detector is used to initialize the end-to-end detector. In the second stage, the backbone and encoder are consistent with the DETR-like model, but only the detector needs to be trained from scratch. Due to this training method, the object detector does not need the additional dataset (ImageNet) to train the backbone, which makes the design of the backbone more flexible and dramatically reduces the training cost of the detector, which is helpful for the practical application of the object detector. At the same time, compared with the DETR-like model, the step-by-step training method can achieve higher accuracy than the traditional training method of the DETR-like model. With the aid of this novel training method, we propose a brand-new end-to-end real-time object detection model called DEYOv3. DEYOv3-N achieves 41.1% on COCO val2017 and 270 FPS on T4 GPU, while DEYOv3-L achieves 51.3% AP and 102 FPS. Without the use of additional training data, DEYOv3 surpasses all existing real-time object detectors in terms of both speed and accuracy. It is worth noting that for models of N, S, and M scales, the training on the COCO dataset can be completed using a single 24GB RTX3090 GPU. Code will be released at https://github.com/ouyanghaodong/DEYOv3.
Lidars and cameras are critical sensors that provide complementary information for 3D detection in autonomous driving. While most prevalent methods progressively downscale the 3D point clouds and camera images and then fuse the high-level features, the downscaled features inevitably lose low-level detailed information. In this paper, we propose Fine-Grained Lidar-Camera Fusion (FGFusion) that make full use of multi-scale features of image and point cloud and fuse them in a fine-grained way. First, we design a dual pathway hierarchy structure to extract both high-level semantic and low-level detailed features of the image. Second, an auxiliary network is introduced to guide point cloud features to better learn the fine-grained spatial information. Finally, we propose multi-scale fusion (MSF) to fuse the last N feature maps of image and point cloud. Extensive experiments on two popular autonomous driving benchmarks, i.e. KITTI and Waymo, demonstrate the effectiveness of our method.
Extreme head postures pose a common challenge across a spectrum of facial analysis tasks, including face detection, facial landmark detection (FLD), and head pose estimation (HPE). These tasks are interdependent, where accurate FLD relies on robust face detection, and HPE is intricately associated with these key points. This paper focuses on the integration of these tasks, particularly when addressing the complexities posed by large-angle face poses. The primary contribution of this study is the proposal of a real-time multi-task detection system capable of simultaneously performing joint detection of faces, facial landmarks, and head poses. This system builds upon the widely adopted YOLOv8 detection framework. It extends the original object detection head by incorporating additional landmark regression head, enabling efficient localization of crucial facial landmarks. Furthermore, we conduct optimizations and enhancements on various modules within the original YOLOv8 framework. To validate the effectiveness and real-time performance of our proposed model, we conduct extensive experiments on 300W-LP and AFLW2000-3D datasets. The results obtained verify the capability of our model to tackle large-angle face pose challenges while delivering real-time performance across these interconnected tasks.
Quantization is a key method for deploying deep neural networks on edge devices with limited memory and computation resources. Recent improvements in Post-Training Quantization (PTQ) methods were achieved by an additional local optimization process for learning the weight quantization rounding policy. However, a gap exists when employing network-wise optimization with small representative datasets. In this paper, we propose a new method for enhanced PTQ (EPTQ) that employs a network-wise quantization optimization process, which benefits from considering cross-layer dependencies during optimization. EPTQ enables network-wise optimization with a small representative dataset using a novel sample-layer attention score based on a label-free Hessian matrix upper bound. The label-free approach makes our method suitable for the PTQ scheme. We give a theoretical analysis for the said bound and use it to construct a knowledge distillation loss that guides the optimization to focus on the more sensitive layers and samples. In addition, we leverage the Hessian upper bound to improve the weight quantization parameters selection by focusing on the more sensitive elements in the weight tensors. Empirically, by employing EPTQ we achieve state-of-the-art results on various models, tasks, and datasets, including ImageNet classification, COCO object detection, and Pascal-VOC for semantic segmentation.
Vision Transformer (ViT) has gained increasing attention in the computer vision community in recent years. However, the core component of ViT, Self-Attention, lacks explicit spatial priors and bears a quadratic computational complexity, thereby constraining the applicability of ViT. To alleviate these issues, we draw inspiration from the recent Retentive Network (RetNet) in the field of NLP, and propose RMT, a strong vision backbone with explicit spatial prior for general purposes. Specifically, we extend the RetNet's temporal decay mechanism to the spatial domain, and propose a spatial decay matrix based on the Manhattan distance to introduce the explicit spatial prior to Self-Attention. Additionally, an attention decomposition form that adeptly adapts to explicit spatial prior is proposed, aiming to reduce the computational burden of modeling global information without disrupting the spatial decay matrix. Based on the spatial decay matrix and the attention decomposition form, we can flexibly integrate explicit spatial prior into the vision backbone with linear complexity. Extensive experiments demonstrate that RMT exhibits exceptional performance across various vision tasks. Specifically, without extra training data, RMT achieves **84.8%** and **86.1%** top-1 acc on ImageNet-1k with **27M/4.5GFLOPs** and **96M/18.2GFLOPs**. For downstream tasks, RMT achieves **54.5** box AP and **47.2** mask AP on the COCO detection task, and **52.8** mIoU on the ADE20K semantic segmentation task. Code is available at https://github.com/qhfan/RMT
The rapid development of 3D object detection systems for self-driving cars has significantly improved accuracy. However, these systems struggle to generalize across diverse driving environments, which can lead to safety-critical failures in detecting traffic participants. To address this, we propose a method that utilizes unlabeled repeated traversals of multiple locations to adapt object detectors to new driving environments. By incorporating statistics computed from repeated LiDAR scans, we guide the adaptation process effectively. Our approach enhances LiDAR-based detection models using spatial quantized historical features and introduces a lightweight regression head to leverage the statistics for feature regularization. Additionally, we leverage the statistics for a novel self-training process to stabilize the training. The framework is detector model-agnostic and experiments on real-world datasets demonstrate significant improvements, achieving up to a 20-point performance gain, especially in detecting pedestrians and distant objects. Code is available at https://github.com/zhangtravis/Hist-DA.
Accurately determining salient regions of an image is challenging when labeled data is scarce. DINO-based self-supervised approaches have recently leveraged meaningful image semantics captured by patch-wise features for locating foreground objects. Recent methods have also incorporated intuitive priors and demonstrated value in unsupervised methods for object partitioning. In this paper, we propose SEMPART, which jointly infers coarse and fine bi-partitions over an image's DINO-based semantic graph. Furthermore, SEMPART preserves fine boundary details using graph-driven regularization and successfully distills the coarse mask semantics into the fine mask. Our salient object detection and single object localization findings suggest that SEMPART produces high-quality masks rapidly without additional post-processing and benefits from co-optimizing the coarse and fine branches.
Recently, the AI community has made significant strides in developing powerful foundation models, driven by large-scale multimodal datasets. However, for audio representation learning, existing datasets suffer from limitations in the following aspects: insufficient volume, simplistic content, and arduous collection procedures. To establish an audio dataset with high-quality captions, we propose an innovative, automatic approach leveraging multimodal inputs, such as video frames, audio streams. Specifically, we construct a large-scale, high-quality, audio-language dataset, named as Auto-ACD, comprising over 1.5M audio-text pairs. We exploit a series of pre-trained models or APIs, to determine audio-visual synchronisation, generate image captions, object detection, or audio tags for specific videos. Subsequently, we employ LLM to paraphrase a congruent caption for each audio, guided by the extracted multi-modality clues. To demonstrate the effectiveness of the proposed dataset, we train widely used models on our dataset and show performance improvement on various downstream tasks, for example, audio-language retrieval, audio captioning, zero-shot classification. In addition, we establish a novel benchmark with environmental information and provide a benchmark for audio-text tasks.
This paper proposes a shape anchor guided learning strategy (AncLearn) for robust holistic indoor scene understanding. We observe that the search space constructed by current methods for proposal feature grouping and instance point sampling often introduces massive noise to instance detection and mesh reconstruction. Accordingly, we develop AncLearn to generate anchors that dynamically fit instance surfaces to (i) unmix noise and target-related features for offering reliable proposals at the detection stage, and (ii) reduce outliers in object point sampling for directly providing well-structured geometry priors without segmentation during reconstruction. We embed AncLearn into a reconstruction-from-detection learning system (AncRec) to generate high-quality semantic scene models in a purely instance-oriented manner. Experiments conducted on the challenging ScanNetv2 dataset demonstrate that our shape anchor-based method consistently achieves state-of-the-art performance in terms of 3D object detection, layout estimation, and shape reconstruction. The code will be available at https://github.com/Geo-Tell/AncRec.
We introduce Dynamic Tiling, a model-agnostic, adaptive, and scalable approach for small object detection, anchored in our inference-data-centric philosophy. Dynamic Tiling starts with non-overlapping tiles for initial detections and utilizes dynamic overlapping rates along with a tile minimizer. This dual approach effectively resolves fragmented objects, improves detection accuracy, and minimizes computational overhead by reducing the number of forward passes through the object detection model. Adaptable to a variety of operational environments, our method negates the need for laborious recalibration. Additionally, our large-small filtering mechanism boosts the detection quality across a range of object sizes. Overall, Dynamic Tiling outperforms existing model-agnostic uniform cropping methods, setting new benchmarks for efficiency and accuracy.