Multi-modality (MM) image fusion aims to render fused images that maintain the merits of different modalities, e.g., functional highlight and detailed textures. To tackle the challenge in modeling cross-modality features and decomposing desirable modality-specific and modality-shared features, we propose a novel Correlation-Driven feature Decomposition Fusion (CDDFuse) network. Firstly, CDDFuse uses Restormer blocks to extract cross-modality shallow features. We then introduce a dual-branch Transformer-CNN feature extractor with Lite Transformer (LT) blocks leveraging long-range attention to handle low-frequency global features and Invertible Neural Networks (INN) blocks focusing on extracting high-frequency local information. A correlation-driven loss is further proposed to make the low-frequency features correlated while the high-frequency features uncorrelated based on the embedded information. Then, the LT-based global fusion and INN-based local fusion layers output the fused image. Extensive experiments demonstrate that our CDDFuse achieves promising results in multiple fusion tasks, including infrared-visible image fusion and medical image fusion. We also show that CDDFuse can boost the performance in downstream infrared-visible semantic segmentation and object detection in a unified benchmark. The code is available at https://github.com/Zhaozixiang1228/MMIF-CDDFuse.
Transformer-based methods have swept the benchmarks on 2D and 3D detection on images. Because tokenization before the attention mechanism drops the spatial information, positional encoding becomes critical for those methods. Recent works found that encodings based on samples of the 3D viewing rays can significantly improve the quality of multi-camera 3D object detection. We hypothesize that 3D point locations can provide more information than rays. Therefore, we introduce 3D point positional encoding, 3DPPE, to the 3D detection Transformer decoder. Although 3D measurements are not available at the inference time of monocular 3D object detection, 3DPPE uses predicted depth to approximate the real point positions. Our hybriddepth module combines direct and categorical depth to estimate the refined depth of each pixel. Despite the approximation, 3DPPE achieves 46.0 mAP and 51.4 NDS on the competitive nuScenes dataset, significantly outperforming encodings based on ray samples. We make the codes available at https://github.com/drilistbox/3DPPE.
Existing object detection methods are bounded in a fixed-set vocabulary by costly labeled data. When dealing with novel categories, the model has to be retrained with more bounding box annotations. Natural language supervision is an attractive alternative for its annotation-free attributes and broader object concepts. However, learning open-vocabulary object detection from language is challenging since image-text pairs do not contain fine-grained object-language alignments. Previous solutions rely on either expensive grounding annotations or distilling classification-oriented vision models. In this paper, we propose a novel open-vocabulary object detection framework directly learning from image-text pair data. We formulate object-language alignment as a set matching problem between a set of image region features and a set of word embeddings. It enables us to train an open-vocabulary object detector on image-text pairs in a much simple and effective way. Extensive experiments on two benchmark datasets, COCO and LVIS, demonstrate our superior performance over the competing approaches on novel categories, e.g. achieving 32.0% mAP on COCO and 21.7% mask mAP on LVIS. Code is available at: https://github.com/clin1223/VLDet.
In this paper, we study the problem of visual grounding by considering both phrase extraction and grounding (PEG). In contrast to the previous phrase-known-at-test setting, PEG requires a model to extract phrases from text and locate objects from images simultaneously, which is a more practical setting in real applications. As phrase extraction can be regarded as a $1$D text segmentation problem, we formulate PEG as a dual detection problem and propose a novel DQ-DETR model, which introduces dual queries to probe different features from image and text for object prediction and phrase mask prediction. Each pair of dual queries is designed to have shared positional parts but different content parts. Such a design effectively alleviates the difficulty of modality alignment between image and text (in contrast to a single query design) and empowers Transformer decoder to leverage phrase mask-guided attention to improve performance. To evaluate the performance of PEG, we also propose a new metric CMAP (cross-modal average precision), analogous to the AP metric in object detection. The new metric overcomes the ambiguity of Recall@1 in many-box-to-one-phrase cases in phrase grounding. As a result, our PEG pre-trained DQ-DETR establishes new state-of-the-art results on all visual grounding benchmarks with a ResNet-101 backbone. For example, it achieves $91.04\%$ and $83.51\%$ in terms of recall rate on RefCOCO testA and testB with a ResNet-101 backbone. Code will be availabl at \url{https://github.com/IDEA-Research/DQ-DETR}.
Few-shot (FS) and zero-shot (ZS) learning are two different approaches for scaling temporal action detection (TAD) to new classes. The former adapts a pretrained vision model to a new task represented by as few as a single video per class, whilst the latter requires no training examples by exploiting a semantic description of the new class. In this work, we introduce a new multi-modality few-shot (MMFS) TAD problem, which can be considered as a marriage of FS-TAD and ZS-TAD by leveraging few-shot support videos and new class names jointly. To tackle this problem, we further introduce a novel MUlti-modality PromPt mETa-learning (MUPPET) method. This is enabled by efficiently bridging pretrained vision and language models whilst maximally reusing already learned capacity. Concretely, we construct multi-modal prompts by mapping support videos into the textual token space of a vision-language model using a meta-learned adapter-equipped visual semantics tokenizer. To tackle large intra-class variation, we further design a query feature regulation scheme. Extensive experiments on ActivityNetv1.3 and THUMOS14 demonstrate that our MUPPET outperforms state-of-the-art alternative methods, often by a large margin. We also show that our MUPPET can be easily extended to tackle the few-shot object detection problem and again achieves the state-of-the-art performance on MS-COCO dataset. The code will be available in https://github.com/sauradip/MUPPET
Fusing data from cameras and LiDAR sensors is an essential technique to achieve robust 3D object detection. One key challenge in camera-LiDAR fusion involves mitigating the large domain gap between the two sensors in terms of coordinates and data distribution when fusing their features. In this paper, we propose a novel camera-LiDAR fusion architecture called, 3D Dual-Fusion, which is designed to mitigate the gap between the feature representations of camera and LiDAR data. The proposed method fuses the features of the camera-view and 3D voxel-view domain and models their interactions through deformable attention. We redesign the transformer fusion encoder to aggregate the information from the two domains. Two major changes include 1) dual query-based deformable attention to fuse the dual-domain features interactively and 2) 3D local self-attention to encode the voxel-domain queries prior to dual-query decoding. The results of an experimental evaluation show that the proposed camera-LiDAR fusion architecture achieved competitive performance on the KITTI and nuScenes datasets, with state-of-the-art performances in some 3D object detection benchmarks categories.
The evaluation of object detection models is usually performed by optimizing a single metric, e.g. mAP, on a fixed set of datasets, e.g. Microsoft COCO and Pascal VOC. Due to image retrieval and annotation costs, these datasets consist largely of images found on the web and do not represent many real-life domains that are being modelled in practice, e.g. satellite, microscopic and gaming, making it difficult to assert the degree of generalization learned by the model. We introduce the Roboflow-100 (RF100) consisting of 100 datasets, 7 imagery domains, 224,714 images, and 805 class labels with over 11,170 labelling hours. We derived RF100 from over 90,000 public datasets, 60 million public images that are actively being assembled and labelled by computer vision practitioners in the open on the web application Roboflow Universe. By releasing RF100, we aim to provide a semantically diverse, multi-domain benchmark of datasets to help researchers test their model's generalizability with real-life data. RF100 download and benchmark replication are available on GitHub.
Most existing domain adaptation (DA) methods align the features based on the domain feature distributions and ignore aspects related to fog, background and target objects, rendering suboptimal performance. In our DA framework, we retain the depth and background information during the domain feature alignment. A consistency loss between the generated depth and fog transmission map is introduced to strengthen the retention of the depth information in the aligned features. To address false object features potentially generated during the DA process, we propose an encoder-decoder framework to reconstruct the fog-free background image. This reconstruction loss also reinforces the encoder, i.e., our DA backbone, to minimize false object features.Moreover, we involve our target data in training both our DA module and our detection module in a semi-supervised manner, so that our detection module is also exposed to the unlabeled target data, the type of data used in the testing stage. Using these ideas, our method significantly outperforms the state-of-the-art method (47.6 mAP against the 44.3 mAP on the Foggy Cityscapes dataset), and obtains the best performance on multiple real-image public datasets. Code is available at: https://github.com/VIML-CVDL/Object-Detection-in-Foggy-Scenes
We propose using Masked Auto-Encoder (MAE), a transformer model self-supervisedly trained on image inpainting, for anomaly detection (AD). Assuming anomalous regions are harder to reconstruct compared with normal regions. MAEDAY is the first image-reconstruction-based anomaly detection method that utilizes a pre-trained model, enabling its use for Few-Shot Anomaly Detection (FSAD). We also show the same method works surprisingly well for the novel tasks of Zero-Shot AD (ZSAD) and Zero-Shot Foreign Object Detection (ZSFOD), where no normal samples are available. Code is available at https://github.com/EliSchwartz/MAEDAY .
Learning descriptive 3D features is crucial for understanding 3D scenes with diverse objects and complex structures. However, it is usually unknown whether important geometric attributes and scene context obtain enough emphasis in an end-to-end trained 3D scene understanding network. To guide 3D feature learning toward important geometric attributes and scene context, we explore the help of textual scene descriptions. Given some free-form descriptions paired with 3D scenes, we extract the knowledge regarding the object relationships and object attributes. We then inject the knowledge to 3D feature learning through three classification-based auxiliary tasks. This language-assisted training can be combined with modern object detection and instance segmentation methods to promote 3D semantic scene understanding, especially in a label-deficient regime. Moreover, the 3D feature learned with language assistance is better aligned with the language features, which can benefit various 3D-language multimodal tasks. Experiments on several benchmarks of 3D-only and 3D-language tasks demonstrate the effectiveness of our language-assisted 3D feature learning. Code is available at https://github.com/Asterisci/Language-Assisted-3D.
The use of computer vision for product and assembly quality control is becoming ubiquitous in the manufacturing industry. Lately, it is apparent that machine learning based solutions are outperforming classical computer vision algorithms in terms of performance and robustness. However, a main drawback is that they require sufficiently large and labeled training datasets, which are often not available or too tedious and too time consuming to acquire. This is especially true for low-volume and high-variance manufacturing. Fortunately, in this industry, CAD models of the manufactured or assembled products are available. This paper introduces CAD2Render, a GPU-accelerated synthetic data generator based on the Unity High Definition Render Pipeline (HDRP). CAD2Render is designed to add variations in a modular fashion, making it possible for high customizable data generation, tailored to the needs of the industrial use case at hand. Although CAD2Render is specifically designed for manufacturing use cases, it can be used for other domains as well. We validate CAD2Render by demonstrating state of the art performance in two industrial relevant setups. We demonstrate that the data generated by our approach can be used to train object detection and pose estimation models with a high enough accuracy to direct a robot. The code for CAD2Render is available at https://github.com/EDM-Research/CAD2Render.
Fully convolutional detectors discard the one-to-many assignment and adopt a one-to-one assigning strategy to achieve end-to-end detection but suffer from the slow convergence issue. In this paper, we revisit these two assignment methods and find that bringing one-to-many assignment back to end-to-end fully convolutional detectors helps with model convergence. Based on this observation, we propose {\em \textbf{D}ual \textbf{A}ssignment} for end-to-end fully convolutional de\textbf{TE}ction (DATE). Our method constructs two branches with one-to-many and one-to-one assignment during training and speeds up the convergence of the one-to-one assignment branch by providing more supervision signals. DATE only uses the branch with the one-to-one matching strategy for model inference, which doesn't bring inference overhead. Experimental results show that Dual Assignment gives nontrivial improvements and speeds up model convergence upon OneNet and DeFCN. Code: https://github.com/YiqunChen1999/date.
The quality of training datasets for deep neural networks is a key factor contributing to the accuracy of resulting models. This effect is amplified in difficult tasks such as object detection. Dealing with errors in datasets is often limited to accepting that some fraction of examples are incorrect, estimating their confidence, and either assigning appropriate weights or ignoring uncertain ones during training. In this work, we propose a different approach. We introduce the Confident Learning for Object Detection (CLOD) algorithm for assessing the quality of each label in object detection datasets, identifying missing, spurious, mislabeled, and mislocated bounding boxes and suggesting corrections. By focusing on finding incorrect examples in the training datasets, we can eliminate them at the root. Suspicious bounding boxes can be reviewed to improve the quality of the dataset, leading to better models without further complicating their already complex architectures. The proposed method is able to point out nearly 80% of artificially disturbed bounding boxes with a false positive rate below 0.1. Cleaning the datasets by applying the most confident automatic suggestions improved mAP scores by 16% to 46%, depending on the dataset, without any modifications to the network architectures. This approach shows promising potential in rectifying state-of-the-art object detection datasets.
LiDAR-produced point clouds are the major source for most state-of-the-art 3D object detectors. Yet, small, distant, and incomplete objects with sparse or few points are often hard to detect. We present Sparse2Dense, a new framework to efficiently boost 3D detection performance by learning to densify point clouds in latent space. Specifically, we first train a dense point 3D detector (DDet) with a dense point cloud as input and design a sparse point 3D detector (SDet) with a regular point cloud as input. Importantly, we formulate the lightweight plug-in S2D module and the point cloud reconstruction module in SDet to densify 3D features and train SDet to produce 3D features, following the dense 3D features in DDet. So, in inference, SDet can simulate dense 3D features from regular (sparse) point cloud inputs without requiring dense inputs. We evaluate our method on the large-scale Waymo Open Dataset and the Waymo Domain Adaptation Dataset, showing its high performance and efficiency over the state of the arts.
Modeling spatial relationship in the data remains critical across many different tasks, such as image classification, semantic segmentation and protein structure understanding. Previous works often use a unified solution like relative positional encoding. However, there exists different kinds of spatial relations, including short-range, medium-range and long-range relations, and modeling them separately can better capture the focus of different tasks on the multi-range relations (e.g., short-range relations can be important in instance segmentation, while long-range relations should be upweighted for semantic segmentation). In this work, we introduce the EurNet for Efficient multi-range relational modeling. EurNet constructs the multi-relational graph, where each type of edge corresponds to short-, medium- or long-range spatial interactions. In the constructed graph, EurNet adopts a novel modeling layer, called gated relational message passing (GRMP), to propagate multi-relational information across the data. GRMP captures multiple relations within the data with little extra computational cost. We study EurNets in two important domains for image and protein structure modeling. Extensive experiments on ImageNet classification, COCO object detection and ADE20K semantic segmentation verify the gains of EurNet over the previous SoTA FocalNet. On the EC and GO protein function prediction benchmarks, EurNet consistently surpasses the previous SoTA GearNet. Our results demonstrate the strength of EurNets on modeling spatial multi-relational data from various domains. The implementations of EurNet for image modeling are available at https://github.com/hirl-team/EurNet-Image . The implementations for other applied domains/tasks will be released soon.
We propose integrally pre-trained transformer pyramid network (iTPN), towards jointly optimizing the network backbone and the neck, so that transfer gap between representation models and downstream tasks is minimal. iTPN is born with two elaborated designs: 1) The first pre-trained feature pyramid upon vision transformer (ViT). 2) Multi-stage supervision to the feature pyramid using masked feature modeling (MFM). iTPN is updated to Fast-iTPN, reducing computational memory overhead and accelerating inference through two flexible designs. 1) Token migration: dropping redundant tokens of the backbone while replenishing them in the feature pyramid without attention operations. 2) Token gathering: reducing computation cost caused by global attention by introducing few gathering tokens. The base/large-level Fast-iTPN achieve 88.75%/89.5% top-1 accuracy on ImageNet-1K. With 1x training schedule using DINO, the base/large-level Fast-iTPN achieves 58.4%/58.8% box AP on COCO object detection, and a 57.5%/58.7% mIoU on ADE20K semantic segmentation using MaskDINO. Fast-iTPN can accelerate the inference procedure by up to 70%, with negligible performance loss, demonstrating the potential to be a powerful backbone for downstream vision tasks. The code is available at: github.com/sunsmarterjie/iTPN.
This paper does not attempt to design a state-of-the-art method for visual recognition but investigates a more efficient way to make use of convolutions to encode spatial features. By comparing the design principles of the recent convolutional neural networks ConvNets) and Vision Transformers, we propose to simplify the self-attention by leveraging a convolutional modulation operation. We show that such a simple approach can better take advantage of the large kernels (>=7x7) nested in convolutional layers. We build a family of hierarchical ConvNets using the proposed convolutional modulation, termed Conv2Former. Our network is simple and easy to follow. Experiments show that our Conv2Former outperforms existent popular ConvNets and vision Transformers, like Swin Transformer and ConvNeXt in all ImageNet classification, COCO object detection and ADE20k semantic segmentation.
Recent LSS-based multi-view 3D object detection has made tremendous progress, by processing the features in Brid-Eye-View (BEV) via the convolutional detector. However, the typical convolution ignores the radial symmetry of the BEV features and increases the difficulty of the detector optimization. To preserve the inherent property of the BEV features and ease the optimization, we propose an azimuth-equivariant convolution (AeConv) and an azimuth-equivariant anchor. The sampling grid of AeConv is always in the radial direction, thus it can learn azimuth-invariant BEV features. The proposed anchor enables the detection head to learn predicting azimuth-irrelevant targets. In addition, we introduce a camera-decoupled virtual depth to unify the depth prediction for the images with different camera intrinsic parameters. The resultant detector is dubbed Azimuth-equivariant Detector (AeDet). Extensive experiments are conducted on nuScenes, and AeDet achieves a 62.0% NDS, surpassing the recent multi-view 3D object detectors such as PETRv2 and BEVDepth by a large margin. Project page: https://fcjian.github.io/aedet.
There has been a significant increase in the adoption of technology in cricket recently. This trend has created the problem of duplicate work being done in similar computer vision-based research works. Our research tries to solve one of these problems by segmenting ball deliveries in a cricket broadcast using deep learning models, MobileNet and YOLO, thus enabling researchers to use our work as a dataset for their research. The output from our research can be used by cricket coaches and players to analyze ball deliveries which are played during the match. This paper presents an approach to segment and extract video shots in which only the ball is being delivered. The video shots are a series of continuous frames that make up the whole scene of the video. Object detection models are applied to reach a high level of accuracy in terms of correctly extracting video shots. The proof of concept for building large datasets of video shots for ball deliveries is proposed which paves the way for further processing on those shots for the extraction of semantics. Ball tracking in these video shots is also done using a separate RetinaNet model as a sample of the usefulness of the proposed dataset. The position on the cricket pitch where the ball lands is also extracted by tracking the ball along the y-axis. The video shot is then classified as a full-pitched, good-length or short-pitched delivery.
In a cyber-physical system such as an autonomous vehicle (AV), machine learning (ML) models can be used to navigate and identify objects that may interfere with the vehicle's operation. However, ML models are unlikely to make accurate decisions when presented with data outside their training distribution. Out-of-distribution (OOD) detection can act as a safety monitor for ML models by identifying such samples at run time. However, in safety critical systems like AVs, OOD detection needs to satisfy real-time constraints in addition to functional requirements. In this demonstration, we use a mobile robot as a surrogate for an AV and use an OOD detector to identify potentially hazardous samples. The robot navigates a miniature town using image data and a YOLO object detection network. We show that our OOD detector is capable of identifying OOD images in real-time on an embedded platform concurrently performing object detection and lane following. We also show that it can be used to successfully stop the vehicle in the presence of unknown, novel samples.