Autonomous systems rely on sensors to estimate the environment around them. However, cameras, LiDARs, and RADARs have their own limitations. In nighttime or degraded environments such as fog, mist, or dust, thermal cameras can provide valuable information regarding the presence of objects of interest due to their heat signature. They make it easy to identify humans and vehicles that are usually at higher temperatures compared to their surroundings. In this paper, we focus on the adaptation of thermal cameras for robotics and automation, where the biggest hurdle is the lack of data. Several multi-modal datasets are available for driving robotics research in tasks such as scene segmentation, object detection, and depth estimation, which are the cornerstone of autonomous systems. However, they are found to be lacking in thermal imagery. Our paper proposes a solution to augment these datasets with synthetic thermal data to enable widespread and rapid adaptation of thermal cameras. We explore the use of conditional diffusion models to convert existing RGB images to thermal images using self-attention to learn the thermal properties of real-world objects.
In Global Navigation Satellite System (GNSS)-denied environments such as indoor parking structures or dense urban canyons, achieving accurate and robust vehicle positioning remains a significant challenge. This paper proposes a cost-effective, vision-based multi-sensor navigation system that integrates monocular depth estimation, semantic filtering, and visual map registration (VMR) with 3-D digital maps. Extensive testing in real-world indoor and outdoor driving scenarios demonstrates the effectiveness of the proposed system, achieving sub-meter accuracy of 92% indoors and more than 80% outdoors, with consistent horizontal positioning and heading average root mean-square errors of approximately 0.98 m and 1.25 {\deg}, respectively. Compared to the baselines examined, the proposed solution significantly reduced drift and improved robustness under various conditions, achieving positioning accuracy improvements of approximately 88% on average. This work highlights the potential of cost-effective monocular vision systems combined with 3D maps for scalable, GNSS-independent navigation in land vehicles.
Accurate 6D object pose estimation is a prerequisite for successfully completing robotic prehensile and non-prehensile manipulation tasks. At present, 6D pose estimation for robotic manipulation generally relies on depth sensors based on, e.g., structured light, time-of-flight, and stereo-vision, which can be expensive, produce noisy output (as compared with RGB cameras), and fail to handle transparent objects. On the other hand, state-of-the-art monocular depth estimation models (MDEMs) provide only affine-invariant depths up to an unknown scale and shift. Metric MDEMs achieve some successful zero-shot results on public datasets, but fail to generalize. We propose a novel framework, Monocular One-shot Metric-depth Alignment (MOMA), to recover metric depth from a single RGB image, through a one-shot adaptation building on MDEM techniques. MOMA performs scale-rotation-shift alignments during camera calibration, guided by sparse ground-truth depth points, enabling accurate depth estimation without additional data collection or model retraining on the testing setup. MOMA supports fine-tuning the MDEM on transparent objects, demonstrating strong generalization capabilities. Real-world experiments on tabletop 2-finger grasping and suction-based bin-picking applications show MOMA achieves high success rates in diverse tasks, confirming its effectiveness.
3D panorama synthesis is a promising yet challenging task that demands high-quality and diverse visual appearance and geometry of the generated omnidirectional content. Existing methods leverage rich image priors from pre-trained 2D foundation models to circumvent the scarcity of 3D panoramic data, but the incompatibility between 3D panoramas and 2D single views limits their effectiveness. In this work, we demonstrate that by applying multi-plane synchronization to the operators from 2D foundation models, their capabilities can be seamlessly extended to the omnidirectional domain. Based on this design, we further introduce DreamCube, a multi-plane RGB-D diffusion model for 3D panorama generation, which maximizes the reuse of 2D foundation model priors to achieve diverse appearances and accurate geometry while maintaining multi-view consistency. Extensive experiments demonstrate the effectiveness of our approach in panoramic image generation, panoramic depth estimation, and 3D scene generation.
This work presents a generalizable framework to transfer relative depth to metric depth. Current monocular depth estimation methods are mainly divided into metric depth estimation (MMDE) and relative depth estimation (MRDE). MMDEs estimate depth in metric scale but are often limited to a specific domain. MRDEs generalize well across different domains, but with uncertain scales which hinders downstream applications. To this end, we aim to build up a framework to solve scale uncertainty and transfer relative depth to metric depth. Previous methods used language as input and estimated two factors for conducting rescaling. Our approach, TR2M, utilizes both text description and image as inputs and estimates two rescale maps to transfer relative depth to metric depth at pixel level. Features from two modalities are fused with a cross-modality attention module to better capture scale information. A strategy is designed to construct and filter confident pseudo metric depth for more comprehensive supervision. We also develop scale-oriented contrastive learning to utilize depth distribution as guidance to enforce the model learning about intrinsic knowledge aligning with the scale distribution. TR2M only exploits a small number of trainable parameters to train on datasets in various domains and experiments not only demonstrate TR2M's great performance in seen datasets but also reveal superior zero-shot capabilities on five unseen datasets. We show the huge potential in pixel-wise transferring relative depth to metric depth with language assistance. (Code is available at: https://github.com/BeileiCui/TR2M)
Dense matching methods like DUSt3R regress pairwise pointmaps for 3D reconstruction. However, the reliance on pairwise prediction and the limited generalization capability inherently restrict the global geometric consistency. In this work, we introduce Test3R, a surprisingly simple test-time learning technique that significantly boosts geometric accuracy. Using image triplets ($I_1,I_2,I_3$), Test3R generates reconstructions from pairs ($I_1,I_2$) and ($I_1,I_3$). The core idea is to optimize the network at test time via a self-supervised objective: maximizing the geometric consistency between these two reconstructions relative to the common image $I_1$. This ensures the model produces cross-pair consistent outputs, regardless of the inputs. Extensive experiments demonstrate that our technique significantly outperforms previous state-of-the-art methods on the 3D reconstruction and multi-view depth estimation tasks. Moreover, it is universally applicable and nearly cost-free, making it easily applied to other models and implemented with minimal test-time training overhead and parameter footprint. Code is available at https://github.com/nopQAQ/Test3R.
Transforming casually captured, monocular videos into fully immersive dynamic experiences is a highly ill-posed task, and comes with significant challenges, e.g., reconstructing unseen regions, and dealing with the ambiguity in monocular depth estimation. In this work we introduce BulletGen, an approach that takes advantage of generative models to correct errors and complete missing information in a Gaussian-based dynamic scene representation. This is done by aligning the output of a diffusion-based video generation model with the 4D reconstruction at a single frozen "bullet-time" step. The generated frames are then used to supervise the optimization of the 4D Gaussian model. Our method seamlessly blends generative content with both static and dynamic scene components, achieving state-of-the-art results on both novel-view synthesis, and 2D/3D tracking tasks.
Vision transformers have been widely explored in various vision tasks. Due to heavy computational cost, much interest has aroused for compressing vision transformer dynamically in the aspect of tokens. Current methods mainly pay attention to token pruning or merging to reduce token numbers, in which tokens are compressed exclusively, causing great information loss and therefore post-training is inevitably required to recover the performance. In this paper, we rethink token reduction and unify the process as an explicit form of token matrix transformation, in which all existing methods are constructing special forms of matrices within the framework. Furthermore, we propose a many-to-many Token Transforming framework that serves as a generalization of all existing methods and reserves the most information, even enabling training-free acceleration. We conduct extensive experiments to validate our framework. Specifically, we reduce 40% FLOPs and accelerate DeiT-S by $\times$1.5 with marginal 0.1% accuracy drop. Furthermore, we extend the method to dense prediction tasks including segmentation, object detection, depth estimation, and language model generation. Results demonstrate that the proposed method consistently achieves substantial improvements, offering a better computation-performance trade-off, impressive budget reduction and inference acceleration.
We estimate scene depth from a single defocus-blurred image using the dark channel as a complementary cue, leveraging its ability to capture local statistics and scene structure. Traditional depth-from-defocus (DFD) methods use multiple images with varying apertures or focus. Single-image DFD is underexplored due to its inherent challenges. Few attempts have focused on depth-from-defocus (DFD) from a single defocused image because the problem is underconstrained. Our method uses the relationship between local defocus blur and contrast variations as depth cues to improve scene structure estimation. The pipeline is trained end-to-end with adversarial learning. Experiments on real data demonstrate that incorporating the dark channel prior into single-image DFD provides meaningful depth estimation, validating our approach.
Depth map enhancement using paired high-resolution RGB images offers a cost-effective solution for improving low-resolution depth data from lightweight ToF sensors. Nevertheless, naively adopting a depth estimation pipeline to fuse the two modalities requires groundtruth depth maps for supervision. To address this, we propose a self-supervised learning framework, SelfToF, which generates detailed and scale-aware depth maps. Starting from an image-based self-supervised depth estimation pipeline, we add low-resolution depth as inputs, design a new depth consistency loss, propose a scale-recovery module, and finally obtain a large performance boost. Furthermore, since the ToF signal sparsity varies in real-world applications, we upgrade SelfToF to SelfToF* with submanifold convolution and guided feature fusion. Consequently, SelfToF* maintain robust performance across varying sparsity levels in ToF data. Overall, our proposed method is both efficient and effective, as verified by extensive experiments on the NYU and ScanNet datasets. The code is available at \href{https://github.com/denyingmxd/selftof}{https://github.com/denyingmxd/selftof}.
Depth estimation is crucial for intelligent systems, enabling applications from autonomous navigation to augmented reality. While traditional stereo and active depth sensors have limitations in cost, power, and robustness, dual-pixel (DP) technology, ubiquitous in modern cameras, offers a compelling alternative. This paper introduces DiFuse-Net, a novel modality decoupled network design for disentangled RGB and DP based depth estimation. DiFuse-Net features a window bi-directional parallax attention mechanism (WBiPAM) specifically designed to capture the subtle DP disparity cues unique to smartphone cameras with small aperture. A separate encoder extracts contextual information from the RGB image, and these features are fused to enhance depth prediction. We also propose a Cross-modal Transfer Learning (CmTL) mechanism to utilize large-scale RGB-D datasets in the literature to cope with the limitations of obtaining large-scale RGB-DP-D dataset. Our evaluation and comparison of the proposed method demonstrates its superiority over the DP and stereo-based baseline methods. Additionally, we contribute a new, high-quality, real-world RGB-DP-D training dataset, named Dual-Camera Dual-Pixel (DCDP) dataset, created using our novel symmetric stereo camera hardware setup, stereo calibration and rectification protocol, and AI stereo disparity estimation method.
Three-dimensional digital urban reconstruction from multi-view aerial images is a critical application where deep multi-view stereo (MVS) methods outperform traditional techniques. However, existing methods commonly overlook the key differences between aerial and close-range settings, such as varying depth ranges along epipolar lines and insensitive feature-matching associated with low-detailed aerial images. To address these issues, we propose an Adaptive Depth Range MVS (ADR-MVS), which integrates monocular geometric cues to improve multi-view depth estimation accuracy. The key component of ADR-MVS is the depth range predictor, which generates adaptive range maps from depth and normal estimates using cross-attention discrepancy learning. In the first stage, the range map derived from monocular cues breaks through predefined depth boundaries, improving feature-matching discriminability and mitigating convergence to local optima. In later stages, the inferred range maps are progressively narrowed, ultimately aligning with the cascaded MVS framework for precise depth regression. Moreover, a normal-guided cost aggregation operation is specially devised for aerial stereo images to improve geometric awareness within the cost volume. Finally, we introduce a normal-guided depth refinement module that surpasses existing RGB-guided techniques. Experimental results demonstrate that ADR-MVS achieves state-of-the-art performance on the WHU, LuoJia-MVS, and M\"unchen datasets, while exhibits superior computational complexity.
Efficient simulation of photon registrations in single-photon LiDAR (SPL) is essential for applications such as depth estimation under high-flux conditions, where hardware dead time significantly distorts photon measurements. However, the conventional wisdom is computationally intensive due to their inherently sequential, photon-by-photon processing. In this paper, we propose a learning-based framework that accelerates the simulation process by modeling the photon count and directly predicting the photon registration probability density function (PDF) using an autoencoder (AE). Our method achieves high accuracy in estimating both the total number of registered photons and their temporal distribution, while substantially reducing simulation time. Extensive experiments validate the effectiveness and efficiency of our approach, highlighting its potential to enable fast and accurate SPL simulations for data-intensive imaging tasks in the high-flux regime.
Transparent and reflective objects in everyday environments pose significant challenges for depth sensors due to their unique visual properties, such as specular reflections and light transmission. These characteristics often lead to incomplete or inaccurate depth estimation, which severely impacts downstream geometry-based vision tasks, including object recognition, scene reconstruction, and robotic manipulation. To address the issue of missing depth information in transparent and reflective objects, we propose DCIRNet, a novel multimodal depth completion network that effectively integrates RGB images and depth maps to enhance depth estimation quality. Our approach incorporates an innovative multimodal feature fusion module designed to extract complementary information between RGB images and incomplete depth maps. Furthermore, we introduce a multi-stage supervision and depth refinement strategy that progressively improves depth completion and effectively mitigates the issue of blurred object boundaries. We integrate our depth completion model into dexterous grasping frameworks and achieve a $44\%$ improvement in the grasp success rate for transparent and reflective objects. We conduct extensive experiments on public datasets, where DCIRNet demonstrates superior performance. The experimental results validate the effectiveness of our approach and confirm its strong generalization capability across various transparent and reflective objects.
The burial state of anthropogenic objects on the seafloor provides insight into localized sedimentation dynamics and is also critical for assessing ecological risks, potential pollutant transport, and the viability of recovery or mitigation strategies for hazardous materials such as munitions. Accurate burial depth estimation from remote imagery remains difficult due to partial occlusion, poor visibility, and object degradation. This work introduces a computer vision pipeline, called PoseIDON, which combines deep foundation model features with multiview photogrammetry to estimate six degrees of freedom object pose and the orientation of the surrounding seafloor from ROV video. Burial depth is inferred by aligning CAD models of the objects with observed imagery and fitting a local planar approximation of the seafloor. The method is validated using footage of 54 objects, including barrels and munitions, recorded at a historic ocean dumpsite in the San Pedro Basin. The model achieves a mean burial depth error of approximately 10 centimeters and resolves spatial burial patterns that reflect underlying sediment transport processes. This approach enables scalable, non-invasive mapping of seafloor burial and supports environmental assessment at contaminated sites.
Optical flow estimation is a crucial subfield of computer vision, serving as a foundation for video tasks. However, the real-world robustness is limited by animated synthetic datasets for training. This introduces domain gaps when applied to real-world applications and limits the benefits of scaling up datasets. To address these challenges, we propose \textbf{Flow-Anything}, a large-scale data generation framework designed to learn optical flow estimation from any single-view images in the real world. We employ two effective steps to make data scaling-up promising. First, we convert a single-view image into a 3D representation using advanced monocular depth estimation networks. This allows us to render optical flow and novel view images under a virtual camera. Second, we develop an Object-Independent Volume Rendering module and a Depth-Aware Inpainting module to model the dynamic objects in the 3D representation. These two steps allow us to generate realistic datasets for training from large-scale single-view images, namely \textbf{FA-Flow Dataset}. For the first time, we demonstrate the benefits of generating optical flow training data from large-scale real-world images, outperforming the most advanced unsupervised methods and supervised methods on synthetic datasets. Moreover, our models serve as a foundation model and enhance the performance of various downstream video tasks.
Language provides a natural interface to specify and evaluate performance on visual tasks. To realize this possibility, vision language models (VLMs) must successfully integrate visual and linguistic information. Our work compares VLMs to a direct readout of their visual encoders to understand their ability to integrate across these modalities. Across a series of vision-centric benchmarks (e.g., depth estimation, correspondence), we find that VLMs perform substantially worse than their visual encoders, dropping to near-chance performance. We investigate these results through a series of analyses across the entire VLM: namely 1) the degradation of vision representations, 2) brittleness to task prompt, and 3) the language model's role in solving the task. We find that the bottleneck in performing these vision-centric tasks lies in this third category; VLMs are not effectively using visual information easily accessible throughout the entire model, and they inherit the language priors present in the LLM. Our work helps diagnose the failure modes of open-source VLMs, and presents a series of evaluations useful for future investigations into visual understanding within VLMs.
Foundation models represent the most prominent and recent paradigm shift in artificial intelligence. Foundation models are large models, trained on broad data that deliver high accuracy in many downstream tasks, often without fine-tuning. For this reason, models such as CLIP , DINO or Vision Transfomers (ViT), are becoming the bedrock of many industrial AI-powered applications. However, the reliance on pre-trained foundation models also introduces significant security concerns, as these models are vulnerable to adversarial attacks. Such attacks involve deliberately crafted inputs designed to deceive AI systems, jeopardizing their reliability. This paper studies the vulnerabilities of vision foundation models, focusing specifically on CLIP and ViTs, and explores the transferability of adversarial attacks to downstream tasks. We introduce a novel attack, targeting the structure of transformer-based architectures in a task-agnostic fashion. We demonstrate the effectiveness of our attack on several downstream tasks: classification, captioning, image/text retrieval, segmentation and depth estimation. Code available at:https://github.com/HondamunigePrasannaSilva/attack-attention
Monocular Depth Estimation (MDE) has emerged as a pivotal task in computer vision, supporting numerous real-world applications. However, deploying accurate depth estimation models on resource-limited edge devices, especially Application-Specific Integrated Circuits (ASICs), is challenging due to the high computational and memory demands. Recent advancements in foundational depth estimation deliver impressive results but further amplify the difficulty of deployment on ASICs. To address this, we propose QuartDepth which adopts post-training quantization to quantize MDE models with hardware accelerations for ASICs. Our approach involves quantizing both weights and activations to 4-bit precision, reducing the model size and computation cost. To mitigate the performance degradation, we introduce activation polishing and compensation algorithm applied before and after activation quantization, as well as a weight reconstruction method for minimizing errors in weight quantization. Furthermore, we design a flexible and programmable hardware accelerator by supporting kernel fusion and customized instruction programmability, enhancing throughput and efficiency. Experimental results demonstrate that our framework achieves competitive accuracy while enabling fast inference and higher energy efficiency on ASICs, bridging the gap between high-performance depth estimation and practical edge-device applicability. Code: https://github.com/shawnricecake/quart-depth
Semantic correspondence (SC) aims to establish semantically meaningful matches across different instances of an object category. We illustrate how recent supervised SC methods remain limited in their ability to generalize beyond sparsely annotated training keypoints, effectively acting as keypoint detectors. To address this, we propose a novel approach for learning dense correspondences by lifting 2D keypoints into a canonical 3D space using monocular depth estimation. Our method constructs a continuous canonical manifold that captures object geometry without requiring explicit 3D supervision or camera annotations. Additionally, we introduce SPair-U, an extension of SPair-71k with novel keypoint annotations, to better assess generalization. Experiments not only demonstrate that our model significantly outperforms supervised baselines on unseen keypoints, highlighting its effectiveness in learning robust correspondences, but that unsupervised baselines outperform supervised counterparts when generalized across different datasets.