Automated salient object detection (SOD) plays an increasingly crucial role in many computer vision applications. By reformulating the depth information as supervision rather than as input, depth-supervised convolutional neural networks (CNN) have achieved promising results on both RGB and RGB-D SOD scenarios with the merits of no requirements for extra depth networks and depth inputs in the inference stage. This paper, for the first time, seeks to expand the applicability of depth supervision to the Transformer architecture. Specifically, we develop a Depth-supervised Fusion TRansformer (DFTR), to further improve the accuracy of both RGB and RGB-D SOD. The proposed DFTR involves three primary features: 1) DFTR, to the best of our knowledge, is the first pure Transformer-based model for depth-supervised SOD; 2) A multi-scale feature aggregation (MFA) module is proposed to fully exploit the multi-scale features encoded by the Swin Transformer in a coarse-to-fine manner; 3) To enable bidirectional information flow across different streams of features, a novel multi-stage feature fusion (MFF) module is further integrated into our DFTR with the emphasis on salient regions at different network learning stages. We extensively evaluate the proposed DFTR on ten benchmarking datasets. Experimental results show that our DFTR consistently outperforms the existing state-of-the-art methods for both RGB and RGB-D SOD tasks. The code and model will be made publicly available.
Multi-object tracking (MOT) aims at estimating bounding boxes and identities of objects in videos. Most methods can be roughly classified as tracking-by-detection and joint-detection-association paradigms. Although the latter has elicited more attention and demonstrates comparable performance relative than the former, we claim that the tracking-by-detection paradigm is still the optimal solution in terms of tracking accuracy,such as ByteTrack,which achieves 80.3 MOTA, 77.3 IDF1 and 63.1 HOTA on the test set of MOT17 with 30 FPS running speed on a single V100 GPU.However, under complex perspectives such as vehicle and UAV acceleration, the performance of such a tracker using uniform Kalman filter will be greatly affected, resulting in tracking loss.In this paper, we propose a variable speed Kalman filter algorithm based on environmental feedback and improve the matching process, which can greatly improve the tracking effect in complex variable speed scenes while maintaining high tracking accuracy in relatively static scenes. Eventually, higher MOTA and IDF1 results can be achieved on MOT17 test set than ByteTrack
The key of referring expression comprehension lies in capturing the cross-modal visual-linguistic relevance. Existing works typically model the cross-modal relevance in each image, where the anchor object/expression and their positive expression/object have the same attribute as the negative expression/object, but with different attribute values. These objects/expressions are exclusively utilized to learn the implicit representation of the attribute by a pair of different values, which however impedes the accuracies of the attribute representations, expression/object representations, and their cross-modal relevances since each anchor object/expression usually has multiple attributes while each attribute usually has multiple potential values. To this end, we investigate a novel REC problem named Group-based REC, where each object/expression is simultaneously employed to construct the multiple triplets among the semantically similar images. To tackle the explosion of the negatives and the differentiation of the anchor-negative relevance scores, we propose the multi-group self-paced relevance learning schema to adaptively assign within-group object-expression pairs with different priorities based on their cross-modal relevances. Since the average cross-modal relevance varies a lot across different groups, we further design an across-group relevance constraint to balance the bias of the group priority. Experiments on three standard REC benchmarks demonstrate the effectiveness and superiority of our method.
We propose a new video camouflaged object detection (VCOD) framework that can exploit both short-term dynamics and long-term temporal consistency to detect camouflaged objects from video frames. An essential property of camouflaged objects is that they usually exhibit patterns similar to the background and thus make them hard to identify from still images. Therefore, effectively handling temporal dynamics in videos becomes the key for the VCOD task as the camouflaged objects will be noticeable when they move. However, current VCOD methods often leverage homography or optical flows to represent motions, where the detection error may accumulate from both the motion estimation error and the segmentation error. On the other hand, our method unifies motion estimation and object segmentation within a single optimization framework. Specifically, we build a dense correlation volume to implicitly capture motions between neighbouring frames and utilize the final segmentation supervision to optimize the implicit motion estimation and segmentation jointly. Furthermore, to enforce temporal consistency within a video sequence, we jointly utilize a spatio-temporal transformer to refine the short-term predictions. Extensive experiments on VCOD benchmarks demonstrate the architectural effectiveness of our approach. We also provide a large-scale VCOD dataset named MoCA-Mask with pixel-level handcrafted ground-truth masks and construct a comprehensive VCOD benchmark with previous methods to facilitate research in this direction. Dataset Link: https://xueliancheng.github.io/SLT-Net-project.
Self-supervised learning (SSL) holds promise in leveraging large amounts of unlabeled data. However, the success of popular SSL methods has limited on single-centric-object images like those in ImageNet and ignores the correlation among the scene and instances, as well as the semantic difference of instances in the scene. To address the above problems, we propose a Unified Self-supervised Visual Pre-training (UniVIP), a novel self-supervised framework to learn versatile visual representations on either single-centric-object or non-iconic dataset. The framework takes into account the representation learning at three levels: 1) the similarity of scene-scene, 2) the correlation of scene-instance, 3) the discrimination of instance-instance. During the learning, we adopt the optimal transport algorithm to automatically measure the discrimination of instances. Massive experiments show that UniVIP pre-trained on non-iconic COCO achieves state-of-the-art transfer performance on a variety of downstream tasks, such as image classification, semi-supervised learning, object detection and segmentation. Furthermore, our method can also exploit single-centric-object dataset such as ImageNet and outperforms BYOL by 2.5% with the same pre-training epochs in linear probing, and surpass current self-supervised object detection methods on COCO dataset, demonstrating its universality and potential.
Recently, adversarial attacks have been applied in visual object tracking to deceive deep trackers by injecting imperceptible perturbations into video frames. However, previous work only generates the video-specific perturbations, which restricts its application scenarios. In addition, existing attacks are difficult to implement in reality due to the real-time of tracking and the re-initialization mechanism. To address these issues, we propose an offline universal adversarial attack called Efficient Universal Shuffle Attack. It takes only one perturbation to cause the tracker malfunction on all videos. To improve the computational efficiency and attack performance, we propose a greedy gradient strategy and a triple loss to efficiently capture and attack model-specific feature representations through the gradients. Experimental results show that EUSA can significantly reduce the performance of state-of-the-art trackers on OTB2015 and VOT2018.
The recently developed DEtection TRansformer (DETR) establishes a new object detection paradigm by eliminating a series of hand-crafted components. However, DETR suffers from extremely slow convergence, which increases the training cost significantly. We observe that the slow convergence is largely attributed to the complication in matching object queries with target features in different feature embedding spaces. This paper presents SAM-DETR, a Semantic-Aligned-Matching DETR that greatly accelerates DETR's convergence without sacrificing its accuracy. SAM-DETR addresses the convergence issue from two perspectives. First, it projects object queries into the same embedding space as encoded image features, where the matching can be accomplished efficiently with aligned semantics. Second, it explicitly searches salient points with the most discriminative features for semantic-aligned matching, which further speeds up the convergence and boosts detection accuracy as well. Being like a plug and play, SAM-DETR complements existing convergence solutions well yet only introduces slight computational overhead. Extensive experiments show that the proposed SAM-DETR achieves superior convergence as well as competitive detection accuracy. The implementation codes are available at https://github.com/ZhangGongjie/SAM-DETR.
3D object detection plays a fundamental role in enabling autonomous driving, which is regarded as the significant key to unlocking the bottleneck of contemporary transportation systems from the perspectives of safety, mobility, and sustainability. Most of the state-of-the-art (SOTA) object detection methods from point clouds are developed based on a single onboard LiDAR, whose performance will be inevitably limited by the range and occlusion, especially in dense traffic scenarios. In this paper, we propose \textit{PillarGrid}, a novel cooperative perception method fusing information from multiple 3D LiDARs (both on-board and roadside), to enhance the situation awareness for connected and automated vehicles (CAVs). PillarGrid consists of four main phases: 1) cooperative preprocessing of point clouds, 2) pillar-wise voxelization and feature extraction, 3) grid-wise deep fusion of features from multiple sensors, and 4) convolutional neural network (CNN)-based augmented 3D object detection. A novel cooperative perception platform is developed for model training and testing. Extensive experimentation shows that PillarGrid outperforms the SOTA single-LiDAR-based 3D object detection methods with respect to both accuracy and range by a large margin.
We present a strong baseline that surpasses the performance of previously published methods on the Habitat Challenge task of navigating to a target object in indoor environments. Our method is motivated from primary failure modes of prior state-of-the-art: poor exploration, inaccurate object identification, and agent getting trapped due to imprecise map construction. We make three contributions to mitigate these issues: (i) First, we show that existing map-based methods fail to effectively use semantic clues for exploration. We present a semantic-agnostic exploration strategy (called Stubborn) without any learning that surprisingly outperforms prior work. (ii) We propose a strategy for integrating temporal information to improve object identification. (iii) Lastly, due to inaccurate depth observation the agent often gets trapped in small regions. We develop a multi-scale collision map for obstacle identification that mitigates this issue.
Self-supervision allows learning meaningful representations of natural images, which usually contain one central object. How well does it transfer to multi-entity scenes? We discuss key aspects of learning structured object-centric representations with self-supervision and validate our insights through several experiments on the CLEVR dataset. Regarding the architecture, we confirm the importance of competition for attention-based object discovery, where each image patch is exclusively attended by one object. For training, we show that contrastive losses equipped with matching can be applied directly in a latent space, avoiding pixel-based reconstruction. However, such an optimization objective is sensitive to false negatives (recurring objects) and false positives (matching errors). Careful consideration is thus required around data augmentation and negative sample selection.
Semi-supervised object detection methods are widely used in autonomous driving systems, where only a fraction of objects are labeled. To propagate information from the labeled objects to the unlabeled ones, pseudo-labels for unlabeled objects must be generated. Although pseudo-labels have proven to improve the performance of semi-supervised object detection significantly, the applications of image-based methods to video frames result in numerous miss or false detections using such generated pseudo-labels. In this paper, we propose a new approach, PseudoProp, to generate robust pseudo-labels by leveraging motion continuity in video frames. Specifically, PseudoProp uses a novel bidirectional pseudo-label propagation approach to compensate for misdetection. A feature-based fusion technique is also used to suppress inference noise. Extensive experiments on the large-scale Cityscapes dataset demonstrate that our method outperforms the state-of-the-art semi-supervised object detection methods by 7.4% on mAP75.
Object detection is an algorithm that recognizes and locates the objects in the image and has a wide range of applications in the visual understanding of complex urban scenes. Existing object detection benchmarks mainly focus on a single specific scenario and their annotation attributes are not rich enough, these make the object detection model is not generalized for the smart city scenes. Considering the diversity and complexity of scenes in intelligent city governance, we build a large-scale object detection benchmark for the smart city. Our benchmark contains about 500K images and includes three scenarios: intelligent transportation, intelligent security, and drones. For the complexity of the real scene in the smart city, the diversity of weather, occlusion, and other complex environment diversity attributes of the images in the three scenes are annotated. The characteristics of the benchmark are analyzed and extensive experiments of the current state-of-the-art target detection algorithm are conducted based on our benchmark to show their performance.
Saliency-driven image and video coding for humans has gained importance in the recent past. In this paper, we propose such a saliency-driven coding framework for the video coding for machines task using the latest video coding standard Versatile Video Coding (VVC). To determine the salient regions before encoding, we employ the real-time-capable object detection network You Only Look Once~(YOLO) in combination with a novel decision criterion. To measure the coding quality for a machine, the state-of-the-art object segmentation network Mask R-CNN was applied to the decoded frame. From extensive simulations we find that, compared to the reference VVC with a constant quality, up to 29 % of bitrate can be saved with the same detection accuracy at the decoder side by applying the proposed saliency-driven framework. Besides, we compare YOLO against other, more traditional saliency detection methods.
Human-Object Interaction (HOI) recognition is challenging due to two factors: (1) significant imbalance across classes and (2) requiring multiple labels per image. This paper shows that these two challenges can be effectively addressed by improving the classifier with the backbone architecture untouched. Firstly, we encode the semantic correlation among classes into the classification head by initializing the weights with language embeddings of HOIs. As a result, the performance is boosted significantly, especially for the few-shot subset. Secondly, we propose a new loss named LSE-Sign to enhance multi-label learning on a long-tailed dataset. Our simple yet effective method enables detection-free HOI classification, outperforming the state-of-the-arts that require object detection and human pose by a clear margin. Moreover, we transfer the classification model to instance-level HOI detection by connecting it with an off-the-shelf object detector. We achieve state-of-the-art without additional fine-tuning.
In this paper, we develop position embedding transformation (PETR) for multi-view 3D object detection. PETR encodes the position information of 3D coordinates into image features, producing the 3D position-aware features. Object query can perceive the 3D position-aware features and perform end-to-end object detection. PETR achieves state-of-the-art performance (50.4% NDS and 44.1% mAP) on standard nuScenes dataset and ranks 1st place on the benchmark. It can serve as a simple yet strong baseline for future research. Code is available at \url{https://github.com/megvii-research/PETR}.
There has been significant progress in image captioning in recent years. The generation of video descriptions is still in its early stages; this is due to the complex nature of videos in comparison to images. Generating paragraph descriptions of a video is even more challenging. Amongst the main issues are temporal object dependencies and complex object-object relationships. Recently, many works are proposed on the generation of multi-sentence video descriptions. The majority of the proposed works are based on a two-step approach: 1) event proposals and 2) caption generation. While these approaches produce good results, they miss out on globally available information. Here we propose the use of global object proposals while generating the video captions. Experimental results on ActivityNet dataset illustrate that the use of global object proposals can produce more informative and correct captions. We also propose three scores to evaluate the object detection capacity of the generator. A qualitative comparison of captions generated by the proposed method and the state-of-the-art techniques proves the efficacy of the proposed method.
Real-world object detection models should be cheap and accurate. Knowledge distillation (KD) can boost the accuracy of a small, cheap detection model by leveraging useful information from a larger teacher model. However, a key challenge is identifying the most informative features produced by the teacher for distillation. In this work, we show that only a very small fraction of features within a ground-truth bounding box are responsible for a teacher's high detection performance. Based on this, we propose Prediction-Guided Distillation (PGD), which focuses distillation on these key predictive regions of the teacher and yields considerable gains in performance over many existing KD baselines. In addition, we propose an adaptive weighting scheme over the key regions to smooth out their influence and achieve even better performance. Our proposed approach outperforms current state-of-the-art KD baselines on a variety of advanced one-stage detection architectures. Specifically, on the COCO dataset, our method achieves between +3.1% and +4.6% AP improvement using ResNet-101 and ResNet-50 as the teacher and student backbones, respectively. On the CrowdHuman dataset, we achieve +3.2% and +2.0% improvements in MR and AP, also using these backbones. Our code is available at https://github.com/ChenhongyiYang/PGD.
We solve object localisation in partial scenes, a new problem of estimating the unknown position of an object (e.g. where is the bag?) given a partial 3D scan of a scene. The proposed solution is based on a novel scene graph model, the Spatial Commonsense Graph (SCG), where objects are the nodes and edges define pairwise distances between them, enriched by concept nodes and relationships from a commonsense knowledge base. This allows SCG to better generalise its spatial inference over unknown 3D scenes. The SCG is used to estimate the unknown position of the target object in two steps: first, we feed the SCG into a novel Proximity Prediction Network, a graph neural network that uses attention to perform distance prediction between the node representing the target object and the nodes representing the observed objects in the SCG; second, we propose a Localisation Module based on circular intersection to estimate the object position using all the predicted pairwise distances in order to be independent of any reference system. We create a new dataset of partially reconstructed scenes to benchmark our method and baselines for object localisation in partial scenes, where our proposed method achieves the best localisation performance.
Robust maritime obstacle detection is essential for fully autonomous unmanned surface vehicles (USVs). The currently widely adopted segmentation-based obstacle detection methods are prone to misclassification of object reflections and sun glitter as obstacles, producing many false positive detections, effectively rendering the methods impractical for USV navigation. However, water-turbulence-induced temporal appearance changes on object reflections are very distinctive from the appearance dynamics of true objects. We harness this property to design WaSR-T, a novel maritime obstacle detection network, that extracts the temporal context from a sequence of recent frames to reduce ambiguity. By learning the local temporal characteristics of object reflection on the water surface, WaSR-T substantially improves obstacle detection accuracy in the presence of reflections and glitter. Compared with existing single-frame methods, WaSR-T reduces the number of false positive detections by 41% overall and by over 53% within the danger zone of the boat, while preserving a high recall, and achieving new state-of-the-art performance on the challenging MODS maritime obstacle detection benchmark. The code, pretrained models and extended datasets are available at https://github.com/lojzezust/WaSR-T
Domain generalisation aims to promote the learning of domain-invariant features while suppressing domain-specific features, so that a model can generalise better to previously unseen target domains. An approach to domain generalisation for object detection is proposed, the first such approach applicable to any object detection architecture. Based on a rigorous mathematical analysis, we extend approaches based on feature alignment with a novel component for performing class conditional alignment at the instance level, in addition to aligning the marginal feature distributions across domains at the image level. This allows us to fully address both components of domain shift, i.e. covariate and concept shift, and learn a domain agnostic feature representation. We perform extensive evaluation with both one-stage (FCOS, YOLO) and two-stage (FRCNN) detectors, on a newly proposed benchmark comprising several different datasets for autonomous driving applications (Cityscapes, BDD10K, ACDC, IDD) as well as the GWHD dataset for precision agriculture, and show consistent improvements to the generalisation and localisation performance over baselines and state-of-the-art.