Recent breakthroughs in large foundation models have enabled the possibility of transferring knowledge pre-trained on vast datasets to domains with limited data availability. Agriculture is one of the domains that lacks sufficient data. This study proposes a framework to train effective, domain-specific, small models from foundation models without manual annotation. Our approach begins with SDM (Segmentation-Description-Matching), a stage that leverages two foundation models: SAM2 (Segment Anything in Images and Videos) for segmentation and OpenCLIP (Open Contrastive Language-Image Pretraining) for zero-shot open-vocabulary classification. In the second stage, a novel knowledge distillation mechanism is utilized to distill compact, edge-deployable models from SDM, enhancing both inference speed and perception accuracy. The complete method, termed SDM-D (Segmentation-Description-Matching-Distilling), demonstrates strong performance across various fruit detection tasks object detection, semantic segmentation, and instance segmentation) without manual annotation. It nearly matches the performance of models trained with abundant labels. Notably, SDM-D outperforms open-set detection methods such as Grounding SAM and YOLO-World on all tested fruit detection datasets. Additionally, we introduce MegaFruits, a comprehensive fruit segmentation dataset encompassing over 25,000 images, and all code and datasets are made publicly available at https://github.com/AgRoboticsResearch/SDM-D.git.
Computer vision tasks such as object detection and segmentation rely on the availability of extensive, accurately annotated datasets. In this work, We present CIA, a modular pipeline, for (1) generating synthetic images for dataset augmentation using Stable Diffusion, (2) filtering out low quality samples using defined quality metrics, (3) forcing the existence of specific patterns in generated images using accurate prompting and ControlNet. In order to show how CIA can be used to search for an optimal augmentation pipeline of training data, we study human object detection in a data constrained scenario, using YOLOv8n on COCO and Flickr30k datasets. We have recorded significant improvement using CIA-generated images, approaching the performances obtained when doubling the amount of real images in the dataset. Our findings suggest that our modular framework can significantly enhance object detection systems, and make it possible for future research to be done on data-constrained scenarios. The framework is available at: github.com/multitel-ai/CIA.
Underwater imagery often suffers from severe degradation resulting in low visual quality and reduced object detection performance. This work aims to evaluate state-of-the-art image enhancement models, investigate their effects on underwater object detection, and explore their potential to improve detection performance. To this end, we apply nine recent underwater image enhancement models, covering physical, non-physical and learning-based categories, to two recent underwater image datasets. Following this, we conduct joint qualitative and quantitative analyses on the original and enhanced images, revealing the discrepancy between the two analyses, and analyzing changes in the quality distribution of the images after enhancement. We then train three recent object detection models on the original datasets, selecting the best-performing detector for further analysis. This detector is subsequently re-trained on the enhanced datasets to evaluate changes in detection performance, highlighting the adverse effect of enhancement on detection performance at the dataset level. Next, we perform a correlation study to examine the relationship between various enhancement metrics and the mean Average Precision (mAP). Finally, we conduct an image-level analysis that reveals images of improved detection performance after enhancement. The findings of this study demonstrate the potential of image enhancement to improve detection performance and provide valuable insights for researchers to further explore the effects of enhancement on detection at the individual image level, rather than at the dataset level. This could enable the selective application of enhancement for improved detection. The data generated, code developed, and supplementary materials are publicly available at: https://github.com/RSSL-MTU/Enhancement-Detection-Analysis.
In the recent years, we have witnessed a paradigm shift in the field of Computer Vision, with the forthcoming of the transformer architecture. Detection Transformers has become a state of the art solution to object detection and is a potential candidate for Road Object Detection in Autonomous Vehicles. Despite the abundance of object detection schemes, real-time DETR models are shown to perform significantly better on inference times, with minimal loss of accuracy and performance. In our work, we used Real-Time DETR (RTDETR) object detection on the BadODD Road Object Detection dataset based in Bangladesh, and performed necessary experimentation and testing. Our results gave a mAP50 score of 0.41518 in the public 60% test set, and 0.28194 in the private 40% test set.
As one of the automotive sensors that have emerged in recent years, 4D millimeter-wave radar has a higher resolution than conventional 3D radar and provides precise elevation measurements. But its point clouds are still sparse and noisy, making it challenging to meet the requirements of autonomous driving. Camera, as another commonly used sensor, can capture rich semantic information. As a result, the fusion of 4D radar and camera can provide an affordable and robust perception solution for autonomous driving systems. However, previous radar-camera fusion methods have not yet been thoroughly investigated, resulting in a large performance gap compared to LiDAR-based methods. Specifically, they ignore the feature-blurring problem and do not deeply interact with image semantic information. To this end, we present a simple but effective multi-stage sampling fusion (MSSF) network based on 4D radar and camera. On the one hand, we design a fusion block that can deeply interact point cloud features with image features, and can be applied to commonly used single-modal backbones in a plug-and-play manner. The fusion block encompasses two types, namely, simple feature fusion (SFF) and multiscale deformable feature fusion (MSDFF). The SFF is easy to implement, while the MSDFF has stronger fusion abilities. On the other hand, we propose a semantic-guided head to perform foreground-background segmentation on voxels with voxel feature re-weighting, further alleviating the problem of feature blurring. Extensive experiments on the View-of-Delft (VoD) and TJ4DRadset datasets demonstrate the effectiveness of our MSSF. Notably, compared to state-of-the-art methods, MSSF achieves a 7.0% and 4.0% improvement in 3D mean average precision on the VoD and TJ4DRadSet datasets, respectively. It even surpasses classical LiDAR-based methods on the VoD dataset.
This paper introduces VisionPAD, a novel self-supervised pre-training paradigm designed for vision-centric algorithms in autonomous driving. In contrast to previous approaches that employ neural rendering with explicit depth supervision, VisionPAD utilizes more efficient 3D Gaussian Splatting to reconstruct multi-view representations using only images as supervision. Specifically, we introduce a self-supervised method for voxel velocity estimation. By warping voxels to adjacent frames and supervising the rendered outputs, the model effectively learns motion cues in the sequential data. Furthermore, we adopt a multi-frame photometric consistency approach to enhance geometric perception. It projects adjacent frames to the current frame based on rendered depths and relative poses, boosting the 3D geometric representation through pure image supervision. Extensive experiments on autonomous driving datasets demonstrate that VisionPAD significantly improves performance in 3D object detection, occupancy prediction and map segmentation, surpassing state-of-the-art pre-training strategies by a considerable margin.
Utilizing temporal information to improve the performance of 3D detection has made great progress recently in the field of autonomous driving. Traditional transformer-based temporal fusion methods suffer from quadratic computational cost and information decay as the length of the frame sequence increases. In this paper, we propose a novel method called MambaDETR, whose main idea is to implement temporal fusion in the efficient state space. Moreover, we design a Motion Elimination module to remove the relatively static objects for temporal fusion. On the standard nuScenes benchmark, our proposed MambaDETR achieves remarkable result in the 3D object detection task, exhibiting state-of-the-art performance among existing temporal fusion methods.
In this paper, we introduce DINO-X, which is a unified object-centric vision model developed by IDEA Research with the best open-world object detection performance to date. DINO-X employs the same Transformer-based encoder-decoder architecture as Grounding DINO 1.5 to pursue an object-level representation for open-world object understanding. To make long-tailed object detection easy, DINO-X extends its input options to support text prompt, visual prompt, and customized prompt. With such flexible prompt options, we develop a universal object prompt to support prompt-free open-world detection, making it possible to detect anything in an image without requiring users to provide any prompt. To enhance the model's core grounding capability, we have constructed a large-scale dataset with over 100 million high-quality grounding samples, referred to as Grounding-100M, for advancing the model's open-vocabulary detection performance. Pre-training on such a large-scale grounding dataset leads to a foundational object-level representation, which enables DINO-X to integrate multiple perception heads to simultaneously support multiple object perception and understanding tasks, including detection, segmentation, pose estimation, object captioning, object-based QA, etc. Experimental results demonstrate the superior performance of DINO-X. Specifically, the DINO-X Pro model achieves 56.0 AP, 59.8 AP, and 52.4 AP on the COCO, LVIS-minival, and LVIS-val zero-shot object detection benchmarks, respectively. Notably, it scores 63.3 AP and 56.5 AP on the rare classes of LVIS-minival and LVIS-val benchmarks, improving the previous SOTA performance by 5.8 AP and 5.0 AP. Such a result underscores its significantly improved capacity for recognizing long-tailed objects.
In many video processing tasks, leveraging large-scale image datasets is a common strategy, as image data is more abundant and facilitates comprehensive knowledge transfer. A typical approach for simulating video from static images involves applying spatial transformations, such as affine transformations and spline warping, to create sequences that mimic temporal progression. However, in tasks like video salient object detection, where both appearance and motion cues are critical, these basic image-to-video techniques fail to produce realistic optical flows that capture the independent motion properties of each object. In this study, we show that image-to-video diffusion models can generate realistic transformations of static images while understanding the contextual relationships between image components. This ability allows the model to generate plausible optical flows, preserving semantic integrity while reflecting the independent motion of scene elements. By augmenting individual images in this way, we create large-scale image-flow pairs that significantly enhance model training. Our approach achieves state-of-the-art performance across all public benchmark datasets, outperforming existing approaches.
Detecting ships in synthetic aperture radar (SAR) images is challenging due to strong speckle noise, complex surroundings, and varying scales. This paper proposes MLDet, a multitask learning framework for SAR ship detection, consisting of object detection, speckle suppression, and target segmentation tasks. An angle classification loss with aspect ratio weighting is introduced to improve detection accuracy by addressing angular periodicity and object proportions. The speckle suppression task uses a dual-feature fusion attention mechanism to reduce noise and fuse shallow and denoising features, enhancing robustness. The target segmentation task, leveraging a rotated Gaussian-mask, aids the network in extracting target regions from cluttered backgrounds and improves detection efficiency with pixel-level predictions. The Gaussian-mask ensures ship centers have the highest probabilities, gradually decreasing outward under a Gaussian distribution. Additionally, a weighted rotated boxes fusion (WRBF) strategy combines multi-direction anchor predictions, filtering anchors beyond boundaries or with high overlap but low confidence. Extensive experiments on SSDD+ and HRSID datasets demonstrate the effectiveness and superiority of MLDet.
Deep neural networks (DNNs) deployed in a cloud often allow users to query models via the APIs. However, these APIs expose the models to model extraction attacks (MEAs). In this attack, the attacker attempts to duplicate the target model by abusing the responses from the API. Backdoor-based DNN watermarking is known as a promising defense against MEAs, wherein the defender injects a backdoor into extracted models via API responses. The backdoor is used as a watermark of the model; if a suspicious model has the watermark (i.e., backdoor), it is verified as an extracted model. This work focuses on object detection (OD) models. Existing backdoor attacks on OD models are not applicable for model watermarking as the defense against MEAs on a realistic threat model. Our proposed approach involves inserting a backdoor into extracted models via APIs by stealthily modifying the bounding-boxes (BBs) of objects detected in queries while keeping the OD capability. In our experiments on three OD datasets, the proposed approach succeeded in identifying the extracted models with 100% accuracy in a wide variety of experimental scenarios.
Current Semi-Supervised Object Detection (SSOD) methods enhance detector performance by leveraging large amounts of unlabeled data, assuming that both labeled and unlabeled data share the same label space. However, in open-set scenarios, the unlabeled dataset contains both in-distribution (ID) classes and out-of-distribution (OOD) classes. Applying semi-supervised detectors in such settings can lead to misclassifying OOD class as ID classes. To alleviate this issue, we propose a simple yet effective method, termed Collaborative Feature-Logits Detector (CFL-Detector). Specifically, we introduce a feature-level clustering method using contrastive loss to clarify vector boundaries in the feature space and highlight class differences. Additionally, by optimizing the logits-level uncertainty classification loss, the model enhances its ability to effectively distinguish between ID and OOD classes. Extensive experiments demonstrate that our method achieves state-of-the-art performance compared to existing methods.
Detecting disasters in underground mining, such as explosions and structural damage, has been a persistent challenge over the years. This problem is compounded for first responders, who often have no clear information about the extent or nature of the damage within the mine. The poor-light or even total darkness inside the mines makes rescue efforts incredibly difficult, leading to a tragic loss of life. In this paper, we propose a novel instance segmentation method called DIS-Mine, specifically designed to identify disaster-affected areas within underground mines under low-light or poor visibility conditions, aiding first responders in rescue efforts. DIS-Mine is capable of detecting objects in images, even in complete darkness, by addressing challenges such as high noise, color distortions, and reduced contrast. The key innovations of DIS-Mine are built upon four core components: i) Image brightness improvement, ii) Instance segmentation with SAM integration, iii) Mask R-CNN-based segmentation, and iv) Mask alignment with feature matching. On top of that, we have collected real-world images from an experimental underground mine, introducing a new dataset named ImageMine, specifically gathered in low-visibility conditions. This dataset serves to validate the performance of DIS-Mine in realistic, challenging environments. Our comprehensive experiments on the ImageMine dataset, as well as on various other datasets demonstrate that DIS-Mine achieves a superior F1 score of 86.0% and mIoU of 72.0%, outperforming state-of-the-art instance segmentation methods, with at least 15x improvement and up to 80% higher precision in object detection.
Cameras can be used to perceive the environment around the vehicle, while affordable radar sensors are popular in autonomous driving systems as they can withstand adverse weather conditions unlike cameras. However, radar point clouds are sparser with low azimuth and elevation resolution that lack semantic and structural information of the scenes, resulting in generally lower radar detection performance. In this work, we directly use the raw range-Doppler (RD) spectrum of radar data, thus avoiding radar signal processing. We independently process camera images within the proposed comprehensive image processing pipeline. Specifically, first, we transform the camera images to Bird's-Eye View (BEV) Polar domain and extract the corresponding features with our camera encoder-decoder architecture. The resultant feature maps are fused with Range-Azimuth (RA) features, recovered from the RD spectrum input from the radar decoder to perform object detection. We evaluate our fusion strategy with other existing methods not only in terms of accuracy but also on computational complexity metrics on RADIal dataset.
Input aggregation is a simple technique used by state-of-the-art LiDAR 3D object detectors to improve detection. However, increasing aggregation is known to have diminishing returns and even performance degradation, due to objects responding differently to the number of aggregated frames. To address this limitation, we propose an efficient adaptive method, which we call Variable Aggregation Detection (VADet). Instead of aggregating the entire scene using a fixed number of frames, VADet performs aggregation per object, with the number of frames determined by an object's observed properties, such as speed and point density. VADet thus reduces the inherent trade-offs of fixed aggregation and is not architecture specific. To demonstrate its benefits, we apply VADet to three popular single-stage detectors and achieve state-of-the-art performance on the Waymo dataset.
Current deep learning approaches in computer vision primarily focus on RGB data sacrificing information. In contrast, RAW images offer richer representation, which is crucial for precise recognition, particularly in challenging conditions like low-light environments. The resultant demand for comprehensive RAW image datasets contrasts with the labor-intensive process of creating specific datasets for individual sensors. To address this, we propose a novel diffusion-based method for generating RAW images guided by RGB images. Our approach integrates an RGB-guidance module for feature extraction from RGB inputs, then incorporates these features into the reverse diffusion process with RGB-guided residual blocks across various resolutions. This approach yields high-fidelity RAW images, enabling the creation of camera-specific RAW datasets. Our RGB2RAW experiments on four DSLR datasets demonstrate state-of-the-art performance. Moreover, RAW-Diffusion demonstrates exceptional data efficiency, achieving remarkable performance with as few as 25 training samples or even fewer. We extend our method to create BDD100K-RAW and Cityscapes-RAW datasets, revealing its effectiveness for object detection in RAW imagery, significantly reducing the amount of required RAW images.
Localizing target objects in images is an important task in computer vision. Often it is the first step towards solving a variety of applications in autonomous driving, maintenance, quality insurance, robotics, and augmented reality. Best in class solutions for this task rely on deep neural networks, which require a set of representative training data for best performance. Creating sets of sufficient quality, variety, and size is often difficult, error prone, and expensive. This is where the method of luminance keying can help: it provides a simple yet effective solution to record high quality data for training object detection and segmentation. We extend previous work that presented luminance keying on the common YCB-V set of household objects by recording the remaining objects of the YCB superset. The additional variety of objects - addition of transparency, multiple color variations, non-rigid objects - further demonstrates the usefulness of luminance keying and might be used to test the applicability of the approach on new 2D object detection and segmentation algorithms.
Existing large video-language models (LVLMs) struggle to comprehend long videos correctly due to limited context. To address this problem, fine-tuning long-context LVLMs and employing GPT-based agents have emerged as promising solutions. However, fine-tuning LVLMs would require extensive high-quality data and substantial GPU resources, while GPT-based agents would rely on proprietary models (e.g., GPT-4o). In this paper, we propose Video Retrieval-Augmented Generation (Video-RAG), a training-free and cost-effective pipeline that employs visually-aligned auxiliary texts to help facilitate cross-modality alignment while providing additional information beyond the visual content. Specifically, we leverage open-source external tools to extract visually-aligned information from pure video data (e.g., audio, optical character, and object detection), and incorporate the extracted information into an existing LVLM as auxiliary texts, alongside video frames and queries, in a plug-and-play manner. Our Video-RAG offers several key advantages: (i) lightweight with low computing overhead due to single-turn retrieval; (ii) easy implementation and compatibility with any LVLM; and (iii) significant, consistent performance gains across long video understanding benchmarks, including Video-MME, MLVU, and LongVideoBench. Notably, our model demonstrates superior performance over proprietary models like Gemini-1.5-Pro and GPT-4o when utilized with a 72B model.
Self-supervised learning has made substantial strides in image processing, while visual pre-training for autonomous driving is still in its infancy. Existing methods often focus on learning geometric scene information while neglecting texture or treating both aspects separately, hindering comprehensive scene understanding. In this context, we are excited to introduce GaussianPretrain, a novel pre-training paradigm that achieves a holistic understanding of the scene by uniformly integrating geometric and texture representations. Conceptualizing 3D Gaussian anchors as volumetric LiDAR points, our method learns a deepened understanding of scenes to enhance pre-training performance with detailed spatial structure and texture, achieving that 40.6% faster than NeRF-based method UniPAD with 70% GPU memory only. We demonstrate the effectiveness of GaussianPretrain across multiple 3D perception tasks, showing significant performance improvements, such as a 7.05% increase in NDS for 3D object detection, boosts mAP by 1.9% in HD map construction and 0.8% improvement on Occupancy prediction. These significant gains highlight GaussianPretrain's theoretical innovation and strong practical potential, promoting visual pre-training development for autonomous driving. Source code will be available at https://github.com/Public-BOTs/GaussianPretrain
Throughout the scientific computing space, deep learning algorithms have shown excellent performance in a wide range of applications. As these deep neural networks (DNNs) continue to mature, the necessary compute required to train them has continued to grow. Today, modern DNNs require millions of FLOPs and days to weeks of training to generate a well-trained model. The training times required for DNNs are oftentimes a bottleneck in DNN research for a variety of deep learning applications, and as such, accelerating and scaling DNN training enables more robust and accelerated research. To that end, in this work, we explore utilizing the NRP Nautilus HyperCluster to automate and scale deep learning model training for three separate applications of DNNs, including overhead object detection, burned area segmentation, and deforestation detection. In total, 234 deep neural models are trained on Nautilus, for a total time of 4,040 hours