Multi-camera 3D object detection aims to detect and localize objects in 3D space using multiple cameras, which has attracted more attention due to its cost-effectiveness trade-off. However, these methods often struggle with the lack of accurate depth estimation caused by the natural weakness of the camera in ranging. Recently, multi-modal fusion and knowledge distillation methods for 3D object detection have been proposed to solve this problem, which are time-consuming during the training phase and not friendly to memory cost. In light of this, we propose PromptDet, a lightweight yet effective 3D object detection framework motivated by the success of prompt learning in 2D foundation model. Our proposed framework, PromptDet, comprises two integral components: a general camera-based detection module, exemplified by models like BEVDet and BEVDepth, and a LiDAR-assisted prompter. The LiDAR-assisted prompter leverages the LiDAR points as a complementary signal, enriched with a minimal set of additional trainable parameters. Notably, our framework is flexible due to our prompt-like design, which can not only be used as a lightweight multi-modal fusion method but also as a camera-only method for 3D object detection during the inference phase. Extensive experiments on nuScenes validate the effectiveness of the proposed PromptDet. As a multi-modal detector, PromptDet improves the mAP and NDS by at most 22.8\% and 21.1\% with fewer than 2\% extra parameters compared with the camera-only baseline. Without LiDAR points, PromptDet still achieves an improvement of at most 2.4\% mAP and 4.0\% NDS with almost no impact on camera detection inference time.
The Long Short-Term Memory (LSTM) networks have traditionally faced challenges in scaling and effectively capturing complex dependencies in visual tasks. The xLSTM architecture has emerged to address these limitations, incorporating exponential gating and a parallel matrix memory structure to enhance performance and scalability. Despite these advancements, the potential of xLSTM in visual computing has not been fully realized, particularly in leveraging autoregressive techniques for improved feature extraction. In this paper, we introduce MAL (Cluster-Masked and Multi-Task Pretraining for Enhanced xLSTM Vision Performance), a novel framework that enhances xLSTM's capabilities through innovative pretraining strategies. We propose a cluster-masked masking method that significantly improves local feature capture and optimizes image scanning efficiency. Additionally, our universal encoder-decoder pretraining approach integrates multiple tasks, including image autoregression, depth estimation, and image segmentation, thereby enhancing the model's adaptability and robustness across diverse visual tasks. Our experimental results demonstrate that MAL surpasses traditional supervised models and fully leverages the scaling potential of xLSTM, setting a new benchmark in visual task performance.
The field of indoor monocular 3D object detection is gaining significant attention, fueled by the increasing demand in VR/AR and robotic applications. However, its advancement is impeded by the limited availability and diversity of 3D training data, owing to the labor-intensive nature of 3D data collection and annotation processes. In this paper, we present V-MIND (Versatile Monocular INdoor Detector), which enhances the performance of indoor 3D detectors across a diverse set of object classes by harnessing publicly available large-scale 2D datasets. By leveraging well-established monocular depth estimation techniques and camera intrinsic predictors, we can generate 3D training data by converting large-scale 2D images into 3D point clouds and subsequently deriving pseudo 3D bounding boxes. To mitigate distance errors inherent in the converted point clouds, we introduce a novel 3D self-calibration loss for refining the pseudo 3D bounding boxes during training. Additionally, we propose a novel ambiguity loss to address the ambiguity that arises when introducing new classes from 2D datasets. Finally, through joint training with existing 3D datasets and pseudo 3D bounding boxes derived from 2D datasets, V-MIND achieves state-of-the-art object detection performance across a wide range of classes on the Omni3D indoor dataset.
In this paper, we study the challenging problem of simultaneously removing haze and estimating depth from real monocular hazy videos. These tasks are inherently complementary: enhanced depth estimation improves dehazing via the atmospheric scattering model (ASM), while superior dehazing contributes to more accurate depth estimation through the brightness consistency constraint (BCC). To tackle these intertwined tasks, we propose a novel depth-centric learning framework that integrates the ASM model with the BCC constraint. Our key idea is that both ASM and BCC rely on a shared depth estimation network. This network simultaneously exploits adjacent dehazed frames to enhance depth estimation via BCC and uses the refined depth cues to more effectively remove haze through ASM. Additionally, we leverage a non-aligned clear video and its estimated depth to independently regularize the dehazing and depth estimation networks. This is achieved by designing two discriminator networks: $D_{MFIR}$ enhances high-frequency details in dehazed videos, and $D_{MDR}$ reduces the occurrence of black holes in low-texture regions. Extensive experiments demonstrate that the proposed method outperforms current state-of-the-art techniques in both video dehazing and depth estimation tasks, especially in real-world hazy scenes. Project page: https://fanjunkai1.github.io/projectpage/DCL/index.html.
Inferring the 3D structure of a scene from a single image is an ill-posed and challenging problem in the field of vision-centric autonomous driving. Existing methods usually employ neural radiance fields to produce voxelized 3D occupancy, lacking instance-level semantic reasoning and temporal photometric consistency. In this paper, we propose ViPOcc, which leverages the visual priors from vision foundation models (VFMs) for fine-grained 3D occupancy prediction. Unlike previous works that solely employ volume rendering for RGB and depth image reconstruction, we introduce a metric depth estimation branch, in which an inverse depth alignment module is proposed to bridge the domain gap in depth distribution between VFM predictions and the ground truth. The recovered metric depth is then utilized in temporal photometric alignment and spatial geometric alignment to ensure accurate and consistent 3D occupancy prediction. Additionally, we also propose a semantic-guided non-overlapping Gaussian mixture sampler for efficient, instance-aware ray sampling, which addresses the redundant and imbalanced sampling issue that still exists in previous state-of-the-art methods. Extensive experiments demonstrate the superior performance of ViPOcc in both 3D occupancy prediction and depth estimation tasks on the KITTI-360 and KITTI Raw datasets. Our code is available at: \url{https://mias.group/ViPOcc}.
In this work, we explore new perspectives on cross-view completion learning by drawing an analogy to self-supervised correspondence learning. Through our analysis, we demonstrate that the cross-attention map within cross-view completion models captures correspondence more effectively than other correlations derived from encoder or decoder features. We verify the effectiveness of the cross-attention map by evaluating on both zero-shot matching and learning-based geometric matching and multi-frame depth estimation. Project page is available at https://cvlab-kaist.github.io/ZeroCo/.
Learning to understand dynamic 3D scenes from imagery is crucial for applications ranging from robotics to scene reconstruction. Yet, unlike other problems where large-scale supervised training has enabled rapid progress, directly supervising methods for recovering 3D motion remains challenging due to the fundamental difficulty of obtaining ground truth annotations. We present a system for mining high-quality 4D reconstructions from internet stereoscopic, wide-angle videos. Our system fuses and filters the outputs of camera pose estimation, stereo depth estimation, and temporal tracking methods into high-quality dynamic 3D reconstructions. We use this method to generate large-scale data in the form of world-consistent, pseudo-metric 3D point clouds with long-term motion trajectories. We demonstrate the utility of this data by training a variant of DUSt3R to predict structure and 3D motion from real-world image pairs, showing that training on our reconstructed data enables generalization to diverse real-world scenes. Project page: https://stereo4d.github.io
The advent of stereoscopic videos has opened new horizons in multimedia, particularly in extended reality (XR) and virtual reality (VR) applications, where immersive content captivates audiences across various platforms. Despite its growing popularity, producing stereoscopic videos remains challenging due to the technical complexities involved in generating stereo parallax. This refers to the positional differences of objects viewed from two distinct perspectives and is crucial for creating depth perception. This complex process poses significant challenges for creators aiming to deliver convincing and engaging presentations. To address these challenges, this paper introduces the Text-driven Stereoscopic Video Generation (T-SVG) system. This innovative, model-agnostic, zero-shot approach streamlines video generation by using text prompts to create reference videos. These videos are transformed into 3D point cloud sequences, which are rendered from two perspectives with subtle parallax differences, achieving a natural stereoscopic effect. T-SVG represents a significant advancement in stereoscopic content creation by integrating state-of-the-art, training-free techniques in text-to-video generation, depth estimation, and video inpainting. Its flexible architecture ensures high efficiency and user-friendliness, allowing seamless updates with newer models without retraining. By simplifying the production pipeline, T-SVG makes stereoscopic video generation accessible to a broader audience, demonstrating its potential to revolutionize the field.
Single-image human mesh recovery is a challenging task due to the ill-posed nature of simultaneous body shape, pose, and camera estimation. Existing estimators work well on images taken from afar, but they break down as the person moves close to the camera. Moreover, current methods fail to achieve both accurate 3D pose and 2D alignment at the same time. Error is mainly introduced by inaccurate perspective projection heuristically derived from orthographic parameters. To resolve this long-standing challenge, we present our method BLADE which accurately recovers perspective parameters from a single image without heuristic assumptions. We start from the inverse relationship between perspective distortion and the person's Z-translation Tz, and we show that Tz can be reliably estimated from the image. We then discuss the important role of Tz for accurate human mesh recovery estimated from close-range images. Finally, we show that, once Tz and the 3D human mesh are estimated, one can accurately recover the focal length and full 3D translation. Extensive experiments on standard benchmarks and real-world close-range images show that our method is the first to accurately recover projection parameters from a single image, and consequently attain state-of-the-art accuracy on 3D pose estimation and 2D alignment for a wide range of images. https://research.nvidia.com/labs/amri/projects/blade/
Image reflection removal is crucial for restoring image quality. Distorted images can negatively impact tasks like object detection and image segmentation. In this paper, we present a novel approach for image reflection removal using a single image. Instead of focusing on model architecture, we introduce a new training technique that can be generalized to image-to-image problems, with input and output being similar in nature. This technique is embodied in our multi-step loss mechanism, which has proven effective in the reflection removal task. Additionally, we address the scarcity of reflection removal training data by synthesizing a high-quality, non-linear synthetic dataset called RefGAN using Pix2Pix GAN. This dataset significantly enhances the model's ability to learn better patterns for reflection removal. We also utilize a ranged depth map, extracted from the depth estimation of the ambient image, as an auxiliary feature, leveraging its property of lacking depth estimations for reflections. Our approach demonstrates superior performance on the SIR^2 benchmark and other real-world datasets, proving its effectiveness by outperforming other state-of-the-art models.
In this work, we present Multiformer, a novel approach to depth-aware video panoptic segmentation (DVPS) based on the mask transformer paradigm. Our method learns object representations that are shared across segmentation, monocular depth estimation, and object tracking subtasks. In contrast to recent unified approaches that progressively refine a common object representation, we propose a hybrid method using task-specific branches within each decoder block, ultimately fusing them into a shared representation at the block interfaces. Extensive experiments on the Cityscapes-DVPS and SemKITTI-DVPS datasets demonstrate that Multiformer achieves state-of-the-art performance across all DVPS metrics, outperforming previous methods by substantial margins. With a ResNet-50 backbone, Multiformer surpasses the previous best result by 3.0 DVPQ points while also improving depth estimation accuracy. Using a Swin-B backbone, Multiformer further improves performance by 4.0 DVPQ points. Multiformer also provides valuable insights into the design of multi-task decoder architectures.
We propose a method for dense depth estimation from an event stream generated when sweeping the focal plane of the driving lens attached to an event camera. In this method, a depth map is inferred from an ``event focal stack'' composed of the event stream using a convolutional neural network trained with synthesized event focal stacks. The synthesized event stream is created from a focal stack generated by Blender for any arbitrary 3D scene. This allows for training on scenes with diverse structures. Additionally, we explored methods to eliminate the domain gap between real event streams and synthetic event streams. Our method demonstrates superior performance over a depth-from-defocus method in the image domain on synthetic and real datasets.
This paper proposes a novel method for omnidirectional 360$\degree$ perception. Most common previous methods relied on equirectangular projection. This representation is easily applicable to 2D operation layers but introduces distortions into the image. Other methods attempted to remove the distortions by maintaining a sphere representation but relied on complicated convolution kernels that failed to show competitive results. In this work, we introduce a transformer-based architecture that, by incorporating a novel ``Spherical Local Self-Attention'' and other spherically-oriented modules, successfully operates in the spherical domain and outperforms the state-of-the-art in 360$\degree$ perception benchmarks for depth estimation and semantic segmentation.
While open-source multi-modal language models perform well on simple question answering tasks, they often fail on complex questions that require multiple capabilities, such as fine-grained recognition, visual grounding, and reasoning, and that demand multi-step solutions. We present TACO, a family of multi-modal large action models designed to improve performance on such complex, multi-step, and multi-modal tasks. During inference, TACO produces chains-of-thought-and-action (CoTA), executes intermediate steps by invoking external tools such as OCR, depth estimation and calculator, then integrates both the thoughts and action outputs to produce coherent responses. To train TACO, we create a large dataset of over 1M synthetic CoTA traces generated with GPT-4o and Python programs. We then experiment with various data filtering and mixing techniques and obtain a final subset of 293K high-quality CoTA examples. This dataset enables TACO to learn complex reasoning and action paths, surpassing existing models trained on instruction tuning data with only direct answers. Our model TACO outperforms the instruction-tuned baseline across 8 benchmarks, achieving a 3.6% improvement on average, with gains of up to 15% in MMVet tasks involving OCR, mathematical reasoning, and spatial reasoning. Training on high-quality CoTA traces sets a new standard for complex multi-modal reasoning, highlighting the need for structured, multi-step instruction tuning in advancing open-source mutli-modal models' capabilities.
Event cameras provide low-latency perception for only milliwatts of power. This makes them highly suitable for resource-restricted, agile robots such as small flying drones. Self-supervised learning based on contrast maximization holds great potential for event-based robot vision, as it foregoes the need for high-frequency ground truth and allows for online learning in the robot's operational environment. However, online, on-board learning raises the major challenge of achieving sufficient computational efficiency for real-time learning, while maintaining competitive visual perception performance. In this work, we improve the time and memory efficiency of the contrast maximization pipeline, making on-device learning of low-latency monocular depth possible. We demonstrate that online learning on board a small drone yields more accurate depth estimates and more successful obstacle avoidance behavior compared to only pre-training. Benchmarking experiments show that the proposed pipeline is not only efficient, but also achieves state-of-the-art depth estimation performance among self-supervised approaches. Our work taps into the unused potential of online, on-device robot learning, promising smaller reality gaps and better performance.
Prior works employing pixel-based Gaussian representation have demonstrated efficacy in feed-forward sparse-view reconstruction. However, such representation necessitates cross-view overlap for accurate depth estimation, and is challenged by object occlusions and frustum truncations. As a result, these methods require scene-centric data acquisition to maintain cross-view overlap and complete scene visibility to circumvent occlusions and truncations, which limits their applicability to scene-centric reconstruction. In contrast, in autonomous driving scenarios, a more practical paradigm is ego-centric reconstruction, which is characterized by minimal cross-view overlap and frequent occlusions and truncations. The limitations of pixel-based representation thus hinder the utility of prior works in this task. In light of this, this paper conducts an in-depth analysis of different representations, and introduces Omni-Gaussian representation with tailored network design to complement their strengths and mitigate their drawbacks. Experiments show that our method significantly surpasses state-of-the-art methods, pixelSplat and MVSplat, in ego-centric reconstruction, and achieves comparable performance to prior works in scene-centric reconstruction.
Event cameras, which feature pixels that independently respond to changes in brightness, are becoming increasingly popular in high-speed applications due to their lower latency, reduced bandwidth requirements, and enhanced dynamic range compared to traditional frame-based cameras. Numerous imaging and vision techniques have leveraged event cameras for high-speed scene understanding by capturing high-framerate, high-dynamic range videos, primarily utilizing the temporal advantages inherent to event cameras. Additionally, imaging and vision techniques have utilized the light field-a complementary dimension to temporal information-for enhanced scene understanding. In this work, we propose "Event Fields", a new approach that utilizes innovative optical designs for event cameras to capture light fields at high speed. We develop the underlying mathematical framework for Event Fields and introduce two foundational frameworks to capture them practically: spatial multiplexing to capture temporal derivatives and temporal multiplexing to capture angular derivatives. To realize these, we design two complementary optical setups one using a kaleidoscope for spatial multiplexing and another using a galvanometer for temporal multiplexing. We evaluate the performance of both designs using a custom-built simulator and real hardware prototypes, showcasing their distinct benefits. Our event fields unlock the full advantages of typical light fields-like post-capture refocusing and depth estimation-now supercharged for high-speed and high-dynamic range scenes. This novel light-sensing paradigm opens doors to new applications in photography, robotics, and AR/VR, and presents fresh challenges in rendering and machine learning.
Generalizing metric monocular depth estimation presents a significant challenge due to its ill-posed nature, while the entanglement between camera parameters and depth amplifies issues further, hindering multi-dataset training and zero-shot accuracy. This challenge is particularly evident in autonomous vehicles and mobile robotics, where data is collected with fixed camera setups, limiting the geometric diversity. Yet, this context also presents an opportunity: the fixed relationship between the camera and the ground plane imposes additional perspective geometry constraints, enabling depth regression via vertical image positions of objects. However, this cue is highly susceptible to overfitting, thus we propose a novel canonical representation that maintains consistency across varied camera setups, effectively disentangling depth from specific parameters and enhancing generalization across datasets. We also propose a novel architecture that adaptively and probabilistically fuses depths estimated via object size and vertical image position cues. A comprehensive evaluation demonstrates the effectiveness of the proposed approach on five autonomous driving datasets, achieving accurate metric depth estimation for varying resolutions, aspect ratios and camera setups. Notably, we achieve comparable accuracy to existing zero-shot methods, despite training on a single dataset with a single-camera setup.
Realtime 4D reconstruction for dynamic scenes remains a crucial challenge for autonomous driving perception. Most existing methods rely on depth estimation through self-supervision or multi-modality sensor fusion. In this paper, we propose Driv3R, a DUSt3R-based framework that directly regresses per-frame point maps from multi-view image sequences. To achieve streaming dense reconstruction, we maintain a memory pool to reason both spatial relationships across sensors and dynamic temporal contexts to enhance multi-view 3D consistency and temporal integration. Furthermore, we employ a 4D flow predictor to identify moving objects within the scene to direct our network focus more on reconstructing these dynamic regions. Finally, we align all per-frame pointmaps consistently to the world coordinate system in an optimization-free manner. We conduct extensive experiments on the large-scale nuScenes dataset to evaluate the effectiveness of our method. Driv3R outperforms previous frameworks in 4D dynamic scene reconstruction, achieving 15x faster inference speed compared to methods requiring global alignment. Code: https://github.com/Barrybarry-Smith/Driv3R.
Existing self-supervised monocular depth estimation (MDE) models attempt to improve nighttime performance by using GANs to transfer nighttime images into their daytime versions. However, this can introduce inconsistencies due to the complexities of real-world daytime lighting variations, which may finally lead to inaccurate estimation results. To address this issue, we leverage physical-prior-knowledge about light wavelength and light attenuation during nighttime. Specifically, our model, Light-Attenuation-Aware Network (LAA-Net), incorporates physical insights from Rayleigh scattering theory for robust nighttime depth estimation: LAA-Net is trained based on red channel values because red light preserves more information under nighttime scenarios due to its longer wavelength. Additionally, based on Beer-Lambert law, we introduce Red Channel Attenuation (RCA) loss to guide LAA-Net's training. Experiments on the RobotCar-Night, nuScenes-Night, RobotCar-Day, and KITTI datasets demonstrate that our model outperforms SOTA models.