General detectors follow the pipeline that feature maps extracted from ConvNets are shared between classification and regression tasks. However, there exists obvious conflicting requirements in multi-orientation object detection that classification is insensitive to orientations, while regression is quite sensitive. To address this issue, we provide an Encoder-Decoder architecture, called Rotated Feature Network (RFN), which produces rotation-sensitive feature maps (RS) for regression and rotation-invariant feature maps (RI) for classification. Specifically, the Encoder unit assigns weights for rotated feature maps. The Decoder unit extracts RS and RI by performing resuming operator on rotated and reweighed feature maps, respectively. To make the rotation-invariant characteristics more reliable, we adopt a metric to quantitatively evaluate the rotation-invariance by adding a constrain item in the loss, yielding a promising detection performance. Compared with the state-of-the-art methods, our method can achieve significant improvement on NWPU VHR-10 and RSOD datasets. We further evaluate the RFN on the scene classification in remote sensing images and object detection in natural images, demonstrating its good generalization ability. The proposed RFN can be integrated into an existing framework, leading to great performance with only a slight increase in model complexity.
We address the problem of 3D object detection from 2D monocular images in autonomous driving scenarios. We propose to lift the 2D images to 3D representations using learned neural networks and leverage existing networks working directly on 3D data to perform 3D object detection and localization. We show that, with carefully designed training mechanism and automatically selected minimally noisy data, such a method is not only feasible, but gives higher results than many methods working on actual 3D inputs acquired from physical sensors. On the challenging KITTI benchmark, we show that our 2D to 3D lifted method outperforms many recent competitive 3D networks while significantly outperforming previous state-of-the-art for 3D detection from monocular images. We also show that a late fusion of the output of the network trained on generated 3D images, with that trained on real 3D images, improves performance. We find the results very interesting and argue that such a method could serve as a highly reliable backup in case of malfunction of expensive 3D sensors, if not potentially making them redundant, at least in the case of low human injury risk autonomous navigation scenarios like warehouse automation.
Object tracking is one of the fundamental problems in visual recognition tasks and has achieved significant improvements in recent years. The achievements often come with the price of enormous hardware consumption and expensive labor effort for consecutive labeling. A missing ingredient for robust tracking is achieving performance with minimal modification on network structure and semi-supervised learning intermittent labeled frames. In this paper, we ad-dress these problems in a Bayesian tracking and detection framework parameterized by neural network outputs. In our framework, the tracking and detection process is formulated in a probabilistic way as multi-objects dynamics and network detection uncertainties. With our formulation, we propose a particle filter-based approximate sampling algorithm for tracking object state estimation. Based on our particle filter inference algorithm, a semi-supervised learn-ing algorithm is utilized for learning tracking network on intermittent labeled frames by variational inference. In our experiments, we provide both mAP and probability-based detection measurements for comparison between our algorithm with non-Bayesian solutions. We also train a semi-supervised tracking network on M2Cai16-Tool-Locations Dataset and compare our results with supervised learning on fully labeled frames.
Inadequate bounding box modeling in regression tasks constrains the performance of one-stage 3D object detection. Our study reveals that the primary reason lies in two aspects: (1) The limited center-offset prediction seriously impairs the bounding box localization since many highest response positions significantly deviate from object centers. (2) The low-quality sample ignored in regression tasks significantly impacts the bounding box prediction since it produces unreliable quality (IoU) rectification. To tackle these problems, we propose Decoupled and Interactive Regression Modeling (DIRM) for one-stage detection. Specifically, Decoupled Attribute Regression (DAR) is implemented to facilitate long regression range modeling for the center attribute through an adaptive multi-sample assignment strategy that deeply decouples bounding box attributes. On the other hand, to enhance the reliability of IoU predictions for low-quality results, Interactive Quality Prediction (IQP) integrates the classification task, proficient in modeling negative samples, with quality prediction for joint optimization. Extensive experiments on Waymo and ONCE datasets demonstrate that DIRM significantly improves the performance of several state-of-the-art methods with minimal additional inference latency. Notably, DIRM achieves state-of-the-art detection performance on both the Waymo and ONCE datasets.
The current trend in computer vision is to utilize one universal model to address all various tasks. Achieving such a universal model inevitably requires incorporating multi-domain data for joint training to learn across multiple problem scenarios. In point cloud based 3D object detection, however, such multi-domain joint training is highly challenging, because large domain gaps among point clouds from different datasets lead to the severe domain-interference problem. In this paper, we propose \textbf{OneDet3D}, a universal one-for-all model that addresses 3D detection across different domains, including diverse indoor and outdoor scenes, within the \emph{same} framework and only \emph{one} set of parameters. We propose the domain-aware partitioning in scatter and context, guided by a routing mechanism, to address the data interference issue, and further incorporate the text modality for a language-guided classification to unify the multi-dataset label spaces and mitigate the category interference issue. The fully sparse structure and anchor-free head further accommodate point clouds with significant scale disparities. Extensive experiments demonstrate the strong universal ability of OneDet3D to utilize only one trained model for addressing almost all 3D object detection tasks.
Service mobile robots are often required to avoid dynamic objects while performing their tasks, but they usually have only limited computational resources. So we present a lightweight multi-modal framework for 3D object detection and trajectory prediction. Our system synergistically integrates LiDAR and camera inputs to achieve real-time perception of pedestrians, vehicles, and riders in 3D space. The framework proposes two novel modules: 1) a Cross-Modal Deformable Transformer (CMDT) for object detection with high accuracy and acceptable amount of computation, and 2) a Reference Trajectory-based Multi-Class Transformer (RTMCT) for efficient and diverse trajectory prediction of mult-class objects with flexible trajectory lengths. Evaluations on the CODa benchmark demonstrate superior performance over existing methods across detection (+2.03% in mAP) and trajectory prediction (-0.408m in minADE5 of pedestrians) metrics. Remarkably, the system exhibits exceptional deployability - when implemented on a wheelchair robot with an entry-level NVIDIA 3060 GPU, it achieves real-time inference at 13.2 fps. To facilitate reproducibility and practical deployment, we release the related code of the method at https://github.com/TossherO/3D_Perception and its ROS inference version at https://github.com/TossherO/ros_packages.
Most recent UAV (Unmanned Aerial Vehicle) detectors focus primarily on general challenge such as uneven distribution and occlusion. However, the neglect of scale challenges, which encompass scale variation and small objects, continues to hinder object detection in UAV images. Although existing works propose solutions, they are implicitly modeled and have redundant steps, so detection performance remains limited. And one specific work addressing the above scale challenges can help improve the performance of UAV image detectors. Compared to natural scenes, scale challenges in UAV images happen with problems of limited perception in comprehensive scales and poor robustness to small objects. We found that complementary learning is beneficial for the detection model to address the scale challenges. Therefore, the paper introduces it to form our scale-robust complementary learning network (SCLNet) in conjunction with the object detection model. The SCLNet consists of two implementations and a cooperation method. In detail, one implementation is based on our proposed scale-complementary decoder and scale-complementary loss function to explicitly extract complementary information as complement, named comprehensive-scale complementary learning (CSCL). Another implementation is based on our proposed contrastive complement network and contrastive complement loss function to explicitly guide the learning of small objects with the rich texture detail information of the large objects, named inter-scale contrastive complementary learning (ICCL). In addition, an end-to-end cooperation (ECoop) between two implementations and with the detection model is proposed to exploit each potential.
Vision-Language Model (VLM) have gained widespread adoption in Open-Vocabulary (OV) object detection and segmentation tasks. Despite they have shown promise on OV-related tasks, their effectiveness in conventional vision tasks has thus far been unevaluated. In this work, we present the systematic review of VLM-based detection and segmentation, view VLM as the foundational model and conduct comprehensive evaluations across multiple downstream tasks for the first time: 1) The evaluation spans eight detection scenarios (closed-set detection, domain adaptation, crowded objects, etc.) and eight segmentation scenarios (few-shot, open-world, small object, etc.), revealing distinct performance advantages and limitations of various VLM architectures across tasks. 2) As for detection tasks, we evaluate VLMs under three finetuning granularities: \textit{zero prediction}, \textit{visual fine-tuning}, and \textit{text prompt}, and further analyze how different finetuning strategies impact performance under varied task. 3) Based on empirical findings, we provide in-depth analysis of the correlations between task characteristics, model architectures, and training methodologies, offering insights for future VLM design. 4) We believe that this work shall be valuable to the pattern recognition experts working in the fields of computer vision, multimodal learning, and vision foundation models by introducing them to the problem, and familiarizing them with the current status of the progress while providing promising directions for future research. A project associated with this review and evaluation has been created at https://github.com/better-chao/perceptual_abilities_evaluation.
Vehicle detection in aerial and satellite images is still challenging due to their tiny appearance in pixels compared to the overall size of remote sensing imagery. Classical methods of object detection very often fail in this scenario due to violation of implicit assumptions made such as rich texture, small to moderate ratios between image size and object size. Satellite video is a very new modality which introduces temporal consistency as inductive bias. Approaches for vehicle detection in satellite video use either background subtraction, frame differencing or subspace methods showing moderate performance (0.26 - 0.82 $F_1$ score). This work proposes to apply recent work on deep learning for wide-area motion imagery (WAMI) on satellite video. We show in a first approach comparable results (0.84 $F_1$) on Planet's SkySat-1 LasVegas video with room for further improvement.
Open-vocabulary 3D Object Detection (OV-3DDet) addresses the detection of objects from an arbitrary list of novel categories in 3D scenes, which remains a very challenging problem. In this work, we propose CoDAv2, a unified framework designed to innovatively tackle both the localization and classification of novel 3D objects, under the condition of limited base categories. For localization, the proposed 3D Novel Object Discovery (3D-NOD) strategy utilizes 3D geometries and 2D open-vocabulary semantic priors to discover pseudo labels for novel objects during training. 3D-NOD is further extended with an Enrichment strategy that significantly enriches the novel object distribution in the training scenes, and then enhances the model's ability to localize more novel objects. The 3D-NOD with Enrichment is termed 3D-NODE. For classification, the Discovery-driven Cross-modal Alignment (DCMA) module aligns features from 3D point clouds and 2D/textual modalities, employing both class-agnostic and class-specific alignments that are iteratively refined to handle the expanding vocabulary of objects. Besides, 2D box guidance boosts the classification accuracy against complex background noises, which is coined as Box-DCMA. Extensive evaluation demonstrates the superiority of CoDAv2. CoDAv2 outperforms the best-performing method by a large margin (AP_Novel of 9.17 vs. 3.61 on SUN-RGBD and 9.12 vs. 3.74 on ScanNetv2). Source code and pre-trained models are available at the GitHub project page.
Few-shot object detection (FSOD) aims to expand an object detector for novel categories given only a few instances for training. The few training samples restrict the performance of FSOD model. Recent text-to-image generation models have shown promising results in generating high-quality images. How applicable these synthetic images are for FSOD tasks remains under-explored. This work extensively studies how synthetic images generated from state-of-the-art text-to-image generators benefit FSOD tasks. We focus on two perspectives: (1) How to use synthetic data for FSOD? (2) How to find representative samples from the large-scale synthetic dataset? We design a copy-paste-based pipeline for using synthetic data. Specifically, saliency object detection is applied to the original generated image, and the minimum enclosing box is used for cropping the main object based on the saliency map. After that, the cropped object is randomly pasted on the image, which comes from the base dataset. We also study the influence of the input text of text-to-image generator and the number of synthetic images used. To construct a representative synthetic training dataset, we maximize the diversity of the selected images via a sample-based and cluster-based method. However, the severe problem of high false positives (FP) ratio of novel categories in FSOD can not be solved by using synthetic data. We propose integrating CLIP, a zero-shot recognition model, into the FSOD pipeline, which can filter 90% of FP by defining a threshold for the similarity score between the detected object and the text of the predicted category. Extensive experiments on PASCAL VOC and MS COCO validate the effectiveness of our method, in which performance gain is up to 21.9% compared to the few-shot baseline.
In an autonomous driving system, it is essential to recognize vehicles, pedestrians and cyclists from images. Besides the high accuracy of the prediction, the requirement of real-time running brings new challenges for convolutional network models. In this report, we introduce a real-time method to detect the 2D objects from images. We aggregate several popular one-stage object detectors and train the models of variety input strategies independently, to yield better performance for accurate multi-scale detection of each category, especially for small objects. For model acceleration, we leverage TensorRT to optimize the inference time of our detection pipeline. As shown in the leaderboard, our proposed detection framework ranks the 2nd place with 75.00% L1 mAP and 69.72% L2 mAP in the real-time 2D detection track of the Waymo Open Dataset Challenges, while our framework achieves the latency of 45.8ms/frame on an Nvidia Tesla V100 GPU.
DAVIS camera, streaming two complementary sensing modalities of asynchronous events and frames, has gradually been used to address major object detection challenges (e.g., fast motion blur and low-light). However, how to effectively leverage rich temporal cues and fuse two heterogeneous visual streams remains a challenging endeavor. To address this challenge, we propose a novel streaming object detector with Transformer, namely SODFormer, which first integrates events and frames to continuously detect objects in an asynchronous manner. Technically, we first build a large-scale multimodal neuromorphic object detection dataset (i.e., PKU-DAVIS-SOD) over 1080.1k manual labels. Then, we design a spatiotemporal Transformer architecture to detect objects via an end-to-end sequence prediction problem, where the novel temporal Transformer module leverages rich temporal cues from two visual streams to improve the detection performance. Finally, an asynchronous attention-based fusion module is proposed to integrate two heterogeneous sensing modalities and take complementary advantages from each end, which can be queried at any time to locate objects and break through the limited output frequency from synchronized frame-based fusion strategies. The results show that the proposed SODFormer outperforms four state-of-the-art methods and our eight baselines by a significant margin. We also show that our unifying framework works well even in cases where the conventional frame-based camera fails, e.g., high-speed motion and low-light conditions. Our dataset and code can be available at https://github.com/dianzl/SODFormer.
In this paper, we propose a novel real-time 6D object pose estimation framework, named G2L-Net. Our network operates on point clouds from RGB-D detection in a divide-and-conquer fashion. Specifically, our network consists of three steps. First, we extract the coarse object point cloud from the RGB-D image by 2D detection. Second, we feed the coarse object point cloud to a translation localization network to perform 3D segmentation and object translation prediction. Third, via the predicted segmentation and translation, we transfer the fine object point cloud into a local canonical coordinate, in which we train a rotation localization network to estimate initial object rotation. In the third step, we define point-wise embedding vector features to capture viewpoint-aware information. To calculate more accurate rotation, we adopt a rotation residual estimator to estimate the residual between initial rotation and ground truth, which can boost initial pose estimation performance. Our proposed G2L-Net is real-time despite the fact multiple steps are stacked via the proposed coarse-to-fine framework. Extensive experiments on two benchmark datasets show that G2L-Net achieves state-of-the-art performance in terms of both accuracy and speed.
Tracking multiple tiny objects is highly challenging due to their weak appearance and limited features. Existing multi-object tracking algorithms generally focus on single-modality scenes, and overlook the complementary characteristics of tiny objects captured by multiple remote sensors. To enhance tracking performance by integrating complementary information from multiple sources, we propose a novel framework called {HGT-Track (Heterogeneous Graph Transformer based Multi-Tiny-Object Tracking)}. Specifically, we first employ a Transformer-based encoder to embed images from different modalities. Subsequently, we utilize Heterogeneous Graph Transformer to aggregate spatial and temporal information from multiple modalities to generate detection and tracking features. Additionally, we introduce a target re-detection module (ReDet) to ensure tracklet continuity by maintaining consistency across different modalities. Furthermore, this paper introduces the first benchmark VT-Tiny-MOT (Visible-Thermal Tiny Multi-Object Tracking) for RGB-T fused multiple tiny object tracking. Extensive experiments are conducted on VT-Tiny-MOT, and the results have demonstrated the effectiveness of our method. Compared to other state-of-the-art methods, our method achieves better performance in terms of MOTA (Multiple-Object Tracking Accuracy) and ID-F1 score. The code and dataset will be made available at https://github.com/xuqingyu26/HGTMT.
In this paper, we propose a novel self-supervised representation learning method, Self-EMD, for object detection. Our method directly trained on unlabeled non-iconic image dataset like COCO, instead of commonly used iconic-object image dataset like ImageNet. We keep the convolutional feature maps as the image embedding to preserve spatial structures and adopt Earth Mover's Distance (EMD) to compute the similarity between two embeddings. Our Faster R-CNN (ResNet50-FPN) baseline achieves 39.8% mAP on COCO, which is on par with the state of the art self-supervised methods pre-trained on ImageNet. More importantly, it can be further improved to 40.4% mAP with more unlabeled images, showing its great potential for leveraging more easily obtained unlabeled data. Code will be made available.
State-of-the-art object detection systems for autonomous driving achieve promising results in clear weather conditions. However, such autonomous safety critical systems also need to work in degrading weather conditions, such as rain, fog and snow. Unfortunately, most approaches evaluate only on the KITTI dataset, which consists only of clear weather scenes. In this paper we address this issue and perform one of the most detailed evaluation on single and dual modality architectures on data captured in real weather conditions. We analyse the performance degradation of these architectures in degrading weather conditions. We demonstrate that an object detection architecture performing good in clear weather might not be able to handle degrading weather conditions. We also perform ablation studies on the dual modality architectures and show their limitations.
In the domain of Image Anomaly Detection (IAD), Existing methods frequently exhibit a paucity of fine-grained, interpretable semantic information, resulting in the detection of anomalous entities or activities that are susceptible to machine illusions. This deficiency often leads to the detection of anomalous entities or actions that are susceptible to machine illusions and lack sufficient explanation. In this thesis, we propose a novel approach to anomaly detection, termed Hoi2Anomaly, which aims to achieve precise discrimination and localization of anomalies. The proposed methodology involves the construction of a multi-modal instruction tuning dataset comprising human-object interaction (HOI) pairs in anomalous scenarios. Second, we have trained an HOI extractor in threat scenarios to localize and match anomalous actions and entities. Finally, explanatory content is generated for the detected anomalous HOI by fine-tuning the visual language pretraining (VLP) framework. The experimental results demonstrate that Hoi2Anomaly surpasses existing generative approaches in terms of precision and explainability. We will release Hoi2Anomaly for the advancement of the field of anomaly detection.
In this thesis we address two related aspects of visual object recognition: the use of motion information, and the use of internal supervision, to help unsupervised learning. These two aspects are inter-related in the current study, since image motion is used for internal supervision, via the detection of spatiotemporal events of active-motion and the use of tracking. Most current work in object recognition deals with static images during both learning and recognition. In contrast, we are interested in a dynamic scene where visual processes, such as detecting motion events and tracking, contribute spatiotemporal information, which is useful for object attention, motion segmentation, 3-D understanding and object interactions. We explore the use of these sources of information in both learning and recognition processes. In the first part of the work, we demonstrate how motion can be used for adaptive detection of object-parts in dynamic environments, while automatically learning new object appearances and poses. In the second and main part of the study we develop methods for using specific types of visual motion to solve two difficult problems in unsupervised visual learning: learning to recognize hands by their appearance and by context, and learning to extract direction of gaze. We use our conclusions in this part to propose a model for several aspects of learning by human infants from their visual environment.
To perform adversarial attacks in the physical world, many studies have proposed adversarial camouflage, a method to hide a target object by applying camouflage patterns on 3D object surfaces. For obtaining optimal physical adversarial camouflage, previous studies have utilized the so-called neural renderer, as it supports differentiability. However, existing neural renderers cannot fully represent various real-world transformations due to a lack of control of scene parameters compared to the legacy photo-realistic renderers. In this paper, we propose the Differentiable Transformation Attack (DTA), a framework for generating a robust physical adversarial pattern on a target object to camouflage it against object detection models with a wide range of transformations. It utilizes our novel Differentiable Transformation Network (DTN), which learns the expected transformation of a rendered object when the texture is changed while preserving the original properties of the target object. Using our attack framework, an adversary can gain both the advantages of the legacy photo-realistic renderers including various physical-world transformations and the benefit of white-box access by offering differentiability. Our experiments show that our camouflaged 3D vehicles can successfully evade state-of-the-art object detection models in the photo-realistic environment (i.e., CARLA on Unreal Engine). Furthermore, our demonstration on a scaled Tesla Model 3 proves the applicability and transferability of our method to the real world.