The current state-of-the-art methods in domain adaptive object detection (DAOD) use Mean Teacher self-labelling, where a teacher model, directly derived as an exponential moving average of the student model, is used to generate labels on the target domain which are then used to improve both models in a positive loop. This couples learning and generating labels on the target domain, and other recent works also leverage the generated labels to add additional domain alignment losses. We believe this coupling is brittle and excessively constrained: there is no guarantee that a student trained only on source data can generate accurate target domain labels and initiate the positive feedback loop, and much better target domain labels can likely be generated by using a large pretrained network that has been exposed to much more data. Vision foundational models are exactly such models, and they have shown impressive task generalization capabilities even when frozen. We want to leverage these models for DAOD and introduce DINO Teacher, which consists of two components. First, we train a new labeller on source data only using a large frozen DINOv2 backbone and show it generates more accurate labels than Mean Teacher. Next, we align the student's source and target image patch features with those from a DINO encoder, driving source and target representations closer to the generalizable DINO representation. We obtain state-of-the-art performance on multiple DAOD datasets. Code available at https://github.com/TRAILab/DINO_Teacher
Maintaining road pavement integrity is crucial for ensuring safe and efficient transportation. Conventional methods for assessing pavement condition are often laborious and susceptible to human error. This paper proposes YOLO9tr, a novel lightweight object detection model for pavement damage detection, leveraging the advancements of deep learning. YOLO9tr is based on the YOLOv9 architecture, incorporating a partial attention block that enhances feature extraction and attention mechanisms, leading to improved detection performance in complex scenarios. The model is trained on a comprehensive dataset comprising road damage images from multiple countries, including an expanded set of damage categories beyond the standard four. This broadened classification range allows for a more accurate and realistic assessment of pavement conditions. Comparative analysis demonstrates YOLO9tr's superior precision and inference speed compared to state-of-the-art models like YOLO8, YOLO9 and YOLO10, achieving a balance between computational efficiency and detection accuracy. The model achieves a high frame rate of up to 136 FPS, making it suitable for real-time applications such as video surveillance and automated inspection systems. The research presents an ablation study to analyze the impact of architectural modifications and hyperparameter variations on model performance, further validating the effectiveness of the partial attention block. The results highlight YOLO9tr's potential for practical deployment in real-time pavement condition monitoring, contributing to the development of robust and efficient solutions for maintaining safe and functional road infrastructure.
3D object detection is an important capability needed in various practical applications such as driver assistance systems. Monocular 3D detection, as a representative general setting among image-based approaches, provides a more economical solution than conventional settings relying on LiDARs but still yields unsatisfactory results. This paper first presents a systematic study on this problem. We observe that the current monocular 3D detection can be simplified as an instance depth estimation problem: The inaccurate instance depth blocks all the other 3D attribute predictions from improving the overall detection performance. Moreover, recent methods directly estimate the depth based on isolated instances or pixels while ignoring the geometric relations across different objects. To this end, we construct geometric relation graphs across predicted objects and use the graph to facilitate depth estimation. As the preliminary depth estimation of each instance is usually inaccurate in this ill-posed setting, we incorporate a probabilistic representation to capture the uncertainty. It provides an important indicator to identify confident predictions and further guide the depth propagation. Despite the simplicity of the basic idea, our method, PGD, obtains significant improvements on KITTI and nuScenes benchmarks, achieving 1st place out of all monocular vision-only methods while still maintaining real-time efficiency. Code and models will be released at https://github.com/open-mmlab/mmdetection3d.
Infrared imaging offers resilience against changing lighting conditions by capturing object temperatures. Yet, in few scenarios, its lack of visual details compared to daytime visible images, poses a significant challenge for human and machine interpretation. This paper proposes a novel diffusion method, dubbed Temporally Consistent Patch Diffusion Models (TC-DPM), for infrared-to-visible video translation. Our method, extending the Patch Diffusion Model, consists of two key components. Firstly, we propose a semantic-guided denoising, leveraging the strong representations of foundational models. As such, our method faithfully preserves the semantic structure of generated visible images. Secondly, we propose a novel temporal blending module to guide the denoising trajectory, ensuring the temporal consistency between consecutive frames. Experiment shows that TC-PDM outperforms state-of-the-art methods by 35.3% in FVD for infrared-to-visible video translation and by 6.1% in AP50 for day-to-night object detection. Our code is publicly available at https://github.com/dzungdoan6/tc-pdm
The recently proposed Detection Transformer (DETR) model successfully applies Transformer to objects detection and achieves comparable performance with two-stage object detection frameworks, such as Faster-RCNN. However, DETR suffers from its slow convergence. Training DETR from scratch needs 500 epochs to achieve a high accuracy. To accelerate its convergence, we propose a simple yet effective scheme for improving the DETR framework, namely Spatially Modulated Co-Attention (SMCA) mechanism. The core idea of SMCA is to conduct location-aware co-attention in DETR by constraining co-attention responses to be high near initially estimated bounding box locations. Our proposed SMCA increases DETR's convergence speed by replacing the original co-attention mechanism in the decoder while keeping other operations in DETR unchanged. Furthermore, by integrating multi-head and scale-selection attention designs into SMCA, our fully-fledged SMCA can achieve better performance compared to DETR with a dilated convolution-based backbone (45.6 mAP at 108 epochs vs. 43.3 mAP at 500 epochs). We perform extensive ablation studies on COCO dataset to validate SMCA. Code is released at https://github.com/gaopengcuhk/SMCA-DETR .
In this paper we evaluate the impact of domain shift on human detection models trained on well known object detection datasets when deployed on data outside the distribution of the training set, as well as propose methods to alleviate such phenomena based on the available annotations from the target domain. Specifically, we introduce the OpenDR Humans in Field dataset, collected in the context of agricultural robotics applications, using the Robotti platform, allowing for quantitatively measuring the impact of domain shift in such applications. Furthermore, we examine the importance of manual annotation by evaluating three distinct scenarios concerning the training data: a) only negative samples, i.e., no depicted humans, b) only positive samples, i.e., only images which contain humans, and c) both negative and positive samples. Our results indicate that good performance can be achieved even when using only negative samples, if additional consideration is given to the training process. We also find that positive samples increase performance especially in terms of better localization. The dataset is publicly available for download at https://github.com/opendr-eu/datasets.
Research into adversarial examples (AE) has developed rapidly, yet static adversarial patches are still the main technique for conducting attacks in the real world, despite being obvious, semi-permanent and unmodifiable once deployed. In this paper, we propose Short-Lived Adversarial Perturbations (SLAP), a novel technique that allows adversaries to realize physically robust real-world AE by using a light projector. Attackers can project a specifically crafted adversarial perturbation onto a real-world object, transforming it into an AE. This allows the adversary greater control over the attack compared to adversarial patches: (i) projections can be dynamically turned on and off or modified at will, (ii) projections do not suffer from the locality constraint imposed by patches, making them harder to detect. We study the feasibility of SLAP in the self-driving scenario, targeting both object detector and traffic sign recognition tasks, focusing on the detection of stop signs. We conduct experiments in a variety of ambient light conditions, including outdoors, showing how in non-bright settings the proposed method generates AE that are extremely robust, causing misclassifications on state-of-the-art networks with up to 99% success rate for a variety of angles and distances. We also demostrate that SLAP-generated AE do not present detectable behaviours seen in adversarial patches and therefore bypass SentiNet, a physical AE detection method. We evaluate other defences including an adaptive defender using adversarial learning which is able to thwart the attack effectiveness up to 80% even in favourable attacker conditions.
Head detection in the indoor video is an essential component of building occupancy detection. While deep models have achieved remarkable progress in general object detection, they are not satisfying enough in complex indoor scenes. The indoor surveillance video often includes cluttered background objects, among which heads have small scales and diverse poses. In this paper, we propose Motion-aware Pseudo Siamese Network (MPSN), an end-to-end approach that leverages head motion information to guide the deep model to extract effective head features in indoor scenarios. By taking the pixel-wise difference of adjacent frames as the auxiliary input, MPSN effectively enhances human head motion information and removes the irrelevant objects in the background. Compared with prior methods, it achieves superior performance on the two indoor video datasets. Our experiments show that MPSN successfully suppresses static background objects and highlights the moving instances, especially human heads in indoor videos. We also compare different methods to capture head motion, which demonstrates the simplicity and flexibility of MPSN. To validate the robustness of MPSN, we conduct adversarial experiments with a mathematical solution of small perturbations for robust model selection. Finally, for confirming its potential in building control systems, we apply MPSN to occupancy counting. Code is available at https://github.com/pl-share/MPSN.
Visual grounding is a task to locate the target indicated by a natural language expression. Existing methods extend the generic object detection framework to this problem. They base the visual grounding on the features from pre-generated proposals or anchors, and fuse these features with the text embeddings to locate the target mentioned by the text. However, modeling the visual features from these predefined locations may fail to fully exploit the visual context and attribute information in the text query, which limits their performance. In this paper, we propose a transformer-based framework for accurate visual grounding by establishing text-conditioned discriminative features and performing multi-stage cross-modal reasoning. Specifically, we develop a visual-linguistic verification module to focus the visual features on regions relevant to the textual descriptions while suppressing the unrelated areas. A language-guided feature encoder is also devised to aggregate the visual contexts of the target object to improve the object's distinctiveness. To retrieve the target from the encoded visual features, we further propose a multi-stage cross-modal decoder to iteratively speculate on the correlations between the image and text for accurate target localization. Extensive experiments on five widely used datasets validate the efficacy of our proposed components and demonstrate state-of-the-art performance. Our code is public at https://github.com/yangli18/VLTVG.
We propose a novel approach to video anomaly detection: we treat feature vectors extracted from videos as realizations of a random variable with a fixed distribution and model this distribution with a neural network. This lets us estimate the likelihood of test videos and detect video anomalies by thresholding the likelihood estimates. We train our video anomaly detector using a modification of denoising score matching, a method that injects training data with noise to facilitate modeling its distribution. To eliminate hyperparameter selection, we model the distribution of noisy video features across a range of noise levels and introduce a regularizer that tends to align the models for different levels of noise. At test time, we combine anomaly indications at multiple noise scales with a Gaussian mixture model. Running our video anomaly detector induces minimal delays as inference requires merely extracting the features and forward-propagating them through a shallow neural network and a Gaussian mixture model. Our experiments on five popular video anomaly detection benchmarks demonstrate state-of-the-art performance, both in the object-centric and in the frame-centric setup.
Current top-performing object detectors depend on deep CNN backbones, such as ResNet-101 and Inception, benefiting from their powerful feature representations but suffering from high computational costs. Conversely, some lightweight model based detectors fulfil real time processing, while their accuracies are often criticized. In this paper, we explore an alternative to build a fast and accurate detector by strengthening lightweight features using a hand-crafted mechanism. Inspired by the structure of Receptive Fields (RFs) in human visual systems, we propose a novel RF Block (RFB) module, which takes the relationship between the size and eccentricity of RFs into account, to enhance the feature discriminability and robustness. We further assemble RFB to the top of SSD, constructing the RFB Net detector. To evaluate its effectiveness, experiments are conducted on two major benchmarks and the results show that RFB Net is able to reach the performance of advanced very deep detectors while keeping the real-time speed. Code is available at https://github.com/ruinmessi/RFBNet.
We introduce Similarity Group Proposal Network (SGPN), a simple and intuitive deep learning framework for 3D object instance segmentation on point clouds. SGPN uses a single network to predict point grouping proposals and a corresponding semantic class for each proposal, from which we can directly extract instance segmentation results. Important to the effectiveness of SGPN is its novel representation of 3D instance segmentation results in the form of a similarity matrix that indicates the similarity between each pair of points in embedded feature space, thus producing an accurate grouping proposal for each point. To the best of our knowledge, SGPN is the first framework to learn 3D instance-aware semantic segmentation on point clouds. Experimental results on various 3D scenes show the effectiveness of our method on 3D instance segmentation, and we also evaluate the capability of SGPN to improve 3D object detection and semantic segmentation results. We also demonstrate its flexibility by seamlessly incorporating 2D CNN features into the framework to boost performance.
Current fine-grained classification approaches often rely on a robust localization of object parts to extract localized feature representations suitable for discrimination. However, part localization is a challenging task due to the large variation of appearance and pose. In this paper, we show how pre-trained convolutional neural networks can be used for robust and efficient object part discovery and localization without the necessity to actually train the network on the current dataset. Our approach called "part detector discovery" (PDD) is based on analyzing the gradient maps of the network outputs and finding activation centers spatially related to annotated semantic parts or bounding boxes. This allows us not just to obtain excellent performance on the CUB200-2011 dataset, but in contrast to previous approaches also to perform detection and bird classification jointly without requiring a given bounding box annotation during testing and ground-truth parts during training. The code is available at http://www.inf-cv.uni-jena.de/part_discovery and https://github.com/cvjena/PartDetectorDisovery.
Current LiDAR-based 3D object detectors for autonomous driving are almost entirely trained on human-annotated data collected in specific geographical domains with specific sensor setups, making it difficult to adapt to a different domain. MODEST is the first work to train 3D object detectors without any labels. Our work, HyperMODEST, proposes a universal method implemented on top of MODEST that can largely accelerate the self-training process and does not require tuning on a specific dataset. We filter intermediate pseudo-labels used for data augmentation with low confidence scores. On the nuScenes dataset, we observe a significant improvement of 1.6% in AP BEV in 0-80m range at IoU=0.25 and an improvement of 1.7% in AP BEV in 0-80m range at IoU=0.5 while only using one-fifth of the training time in the original approach by MODEST. On the Lyft dataset, we also observe an improvement over the baseline during the first round of iterative self-training. We explore the trade-off between high precision and high recall in the early stage of the self-training process by comparing our proposed method with two other score filtering methods: confidence score filtering for pseudo-labels with and without static label retention. The code and models of this work are available at https://github.com/TRAILab/HyperMODEST
Multimodal Large Language Models (MLLMs) like GPT-4V are capable of reasoning across text and image modalities, showing promise in a variety of complex vision-language tasks. In this preliminary study, we investigate the out-of-the-box capabilities of GPT-4V in the domain of image forensics, specifically, in detecting image splicing manipulations. Without any task-specific fine-tuning, we evaluate GPT-4V using three prompting strategies: Zero-Shot (ZS), Few-Shot (FS), and Chain-of-Thought (CoT), applied over a curated subset of the CASIA v2.0 splicing dataset. Our results show that GPT-4V achieves competitive detection performance in zero-shot settings (more than 85% accuracy), with CoT prompting yielding the most balanced trade-off across authentic and spliced images. Qualitative analysis further reveals that the model not only detects low-level visual artifacts but also draws upon real-world contextual knowledge such as object scale, semantic consistency, and architectural facts, to identify implausible composites. While GPT-4V lags behind specialized state-of-the-art splicing detection models, its generalizability, interpretability, and encyclopedic reasoning highlight its potential as a flexible tool in image forensics.
We propose Rank & Sort (RS) Loss, a ranking-based loss function to train deep object detection and instance segmentation methods (i.e. visual detectors). RS Loss supervises the classifier, a sub-network of these methods, to rank each positive above all negatives as well as to sort positives among themselves with respect to (wrt.) their localisation qualities (e.g. Intersection-over-Union - IoU). To tackle the non-differentiable nature of ranking and sorting, we reformulate the incorporation of error-driven update with backpropagation as Identity Update, which enables us to model our novel sorting error among positives. With RS Loss, we significantly simplify training: (i) Thanks to our sorting objective, the positives are prioritized by the classifier without an additional auxiliary head (e.g. for centerness, IoU, mask-IoU), (ii) due to its ranking-based nature, RS Loss is robust to class imbalance, and thus, no sampling heuristic is required, and (iii) we address the multi-task nature of visual detectors using tuning-free task-balancing coefficients. Using RS Loss, we train seven diverse visual detectors only by tuning the learning rate, and show that it consistently outperforms baselines: e.g. our RS Loss improves (i) Faster R-CNN by ~ 3 box AP and aLRP Loss (ranking-based baseline) by ~ 2 box AP on COCO dataset, (ii) Mask R-CNN with repeat factor sampling (RFS) by 3.5 mask AP (~ 7 AP for rare classes) on LVIS dataset; and also outperforms all counterparts. Code is available at: https://github.com/kemaloksuz/RankSortLoss
Pedestrian safety is one primary concern in autonomous driving. The under-representation of vulnerable groups in today's pedestrian datasets points to an urgent need for a dataset of vulnerable road users. In order to help train comprehensive models and subsequently drive research to improve the accuracy of vulnerable pedestrian identification, we first introduce a new dataset for vulnerable pedestrian detection in this paper: the BG Vulnerable Pedestrian (BGVP) dataset. The dataset includes four classes, i.e., Children Without Disability, Elderly without Disability, With Disability, and Non-Vulnerable. This dataset consists of images collected from the public domain and manually-annotated bounding boxes. In addition, on the proposed dataset, we have trained and tested five classic or state-of-the-art object detection models, i.e., YOLOv4, YOLOv5, YOLOX, Faster R-CNN, and EfficientDet. Our results indicate that YOLOX and YOLOv4 perform the best on our dataset, YOLOv4 scoring 0.7999 and YOLOX scoring 0.7779 on the mAP 0.5 metric, while YOLOX outperforms YOLOv4 by 3.8 percent on the mAP 0.5:0.95 metric. Generally speaking, all five detectors do well predicting the With Disability class and perform poorly in the Elderly Without Disability class. YOLOX consistently outperforms all other detectors on the mAP (0.5:0.95) per class metric, obtaining 0.5644, 0.5242, 0.4781, and 0.6796 for Children Without Disability, Elderly Without Disability, Non-vulnerable, and With Disability, respectively. Our dataset and codes are available at https://github.com/devvansh1997/BGVP.
Multi-camera 3D object detection for autonomous driving is a challenging problem that has garnered notable attention from both academia and industry. An obstacle encountered in vision-based techniques involves the precise extraction of geometry-conscious features from RGB images. Recent approaches have utilized geometric-aware image backbones pretrained on depth-relevant tasks to acquire spatial information. However, these approaches overlook the critical aspect of view transformation, resulting in inadequate performance due to the misalignment of spatial knowledge between the image backbone and view transformation. To address this issue, we propose a novel geometric-aware pretraining framework called GAPretrain. Our approach incorporates spatial and structural cues to camera networks by employing the geometric-rich modality as guidance during the pretraining phase. The transference of modal-specific attributes across different modalities is non-trivial, but we bridge this gap by using a unified bird's-eye-view (BEV) representation and structural hints derived from LiDAR point clouds to facilitate the pretraining process. GAPretrain serves as a plug-and-play solution that can be flexibly applied to multiple state-of-the-art detectors. Our experiments demonstrate the effectiveness and generalization ability of the proposed method. We achieve 46.2 mAP and 55.5 NDS on the nuScenes val set using the BEVFormer method, with a gain of 2.7 and 2.1 points, respectively. We also conduct experiments on various image backbones and view transformations to validate the efficacy of our approach. Code will be released at https://github.com/OpenDriveLab/BEVPerception-Survey-Recipe.
The era of vision-language models (VLMs) trained on web-scale datasets challenges conventional formulations of "open-world" perception. In this work, we revisit the task of few-shot object detection (FSOD) in the context of recent foundational VLMs. First, we point out that zero-shot predictions from VLMs such as GroundingDINO significantly outperform state-of-the-art few-shot detectors (48 vs. 33 AP) on COCO. Despite their strong zero-shot performance, such foundation models may still be sub-optimal. For example, trucks on the web may be defined differently from trucks for a target application such as autonomous vehicle perception. We argue that the task of few-shot recognition can be reformulated as aligning foundation models to target concepts using a few examples. Interestingly, such examples can be multi-modal, using both text and visual cues, mimicking instructions that are often given to human annotators when defining a target concept of interest. Concretely, we propose Foundational FSOD, a new benchmark protocol that evaluates detectors pre-trained on any external data and fine-tuned on multi-modal (text and visual) K-shot examples per target class. We repurpose nuImages for Foundational FSOD, benchmark several popular open-source VLMs, and provide an empirical analysis of state-of-the-art methods. Lastly, we discuss our recent CVPR 2024 Foundational FSOD competition and share insights from the community. Notably, the winning team significantly outperforms our baseline by 23.3 mAP! Our code and dataset splits are available at https://github.com/anishmadan23/foundational_fsod
The comic domain is rapidly advancing with the development of single-page analysis and synthesis models. However, evaluation metrics and datasets lag behind, often limited to small-scale or single-style test sets. We introduce a novel benchmark, CoMix, designed to evaluate the multi-task capabilities of models in comic analysis. Unlike existing benchmarks that focus on isolated tasks such as object detection or text recognition, CoMix addresses a broader range of tasks including object detection, speaker identification, character re-identification, reading order, and multi-modal reasoning tasks like character naming and dialogue generation. Our benchmark comprises three existing datasets with expanded annotations to support multi-task evaluation. To mitigate the over-representation of manga-style data, we have incorporated a new dataset of carefully selected American comic-style books, thereby enriching the diversity of comic styles. CoMix is designed to assess pre-trained models in zero-shot and limited fine-tuning settings, probing their transfer capabilities across different comic styles and tasks. The validation split of the benchmark is publicly available for research purposes, and an evaluation server for the held-out test split is also provided. Comparative results between human performance and state-of-the-art models reveal a significant performance gap, highlighting substantial opportunities for advancements in comic understanding. The dataset, baseline models, and code are accessible at https://github.com/emanuelevivoli/CoMix-dataset. This initiative sets a new standard for comprehensive comic analysis, providing the community with a common benchmark for evaluation on a large and varied set.