The resolution of voxel queries significantly influences the quality of view transformation in camera-based 3D occupancy prediction. However, computational constraints and the practical necessity for real-time deployment require smaller query resolutions, which inevitably leads to an information loss. Therefore, it is essential to encode and preserve rich visual details within limited query sizes while ensuring a comprehensive representation of 3D occupancy. To this end, we introduce ProtoOcc, a novel occupancy network that leverages prototypes of clustered image segments in view transformation to enhance low-resolution context. In particular, the mapping of 2D prototypes onto 3D voxel queries encodes high-level visual geometries and complements the loss of spatial information from reduced query resolutions. Additionally, we design a multi-perspective decoding strategy to efficiently disentangle the densely compressed visual cues into a high-dimensional 3D occupancy scene. Experimental results on both Occ3D and SemanticKITTI benchmarks demonstrate the effectiveness of the proposed method, showing clear improvements over the baselines. More importantly, ProtoOcc achieves competitive performance against the baselines even with 75% reduced voxel resolution. Project page: https://kuai-lab.github.io/cvpr2025protoocc.

Discrete Wavelet Transform (DWT) has been widely explored to enhance the performance of image superresolution (SR). Despite some DWT-based methods improving SR by capturing fine-grained frequency signals, most existing approaches neglect the interrelations among multiscale frequency sub-bands, resulting in inconsistencies and unnatural artifacts in the reconstructed images. To address this challenge, we propose a Diffusion Transformer model based on image Wavelet spectra for SR (DTWSR). DTWSR incorporates the superiority of diffusion models and transformers to capture the interrelations among multiscale frequency sub-bands, leading to a more consistence and realistic SR image. Specifically, we use a Multi-level Discrete Wavelet Transform to decompose images into wavelet spectra. A pyramid tokenization method is proposed which embeds the spectra into a sequence of tokens for transformer model, facilitating to capture features from both spatial and frequency domain. A dual-decoder is designed elaborately to handle the distinct variances in low-frequency and high-frequency sub-bands, without omitting their alignment in image generation. Extensive experiments on multiple benchmark datasets demonstrate the effectiveness of our method, with high performance on both perception quality and fidelity.

Navigating in an unfamiliar environment during deployment poses a critical challenge for a vision-language navigation (VLN) agent. Yet, test-time adaptation (TTA) remains relatively underexplored in robotic navigation, leading us to the fundamental question: what are the key properties of TTA for online VLN? In our view, effective adaptation requires three qualities: 1) flexibility in handling different navigation outcomes, 2) interactivity with external environment, and 3) maintaining a harmony between plasticity and stability. To address this, we introduce FeedTTA, a novel TTA framework for online VLN utilizing feedback-based reinforcement learning. Specifically, FeedTTA learns by maximizing binary episodic feedback, a practical setup in which the agent receives a binary scalar after each episode that indicates the success or failure of the navigation. Additionally, we propose a gradient regularization technique that leverages the binary structure of FeedTTA to achieve a balance between plasticity and stability during adaptation. Our extensive experiments on challenging VLN benchmarks demonstrate the superior adaptability of FeedTTA, even outperforming the state-of-the-art offline training methods in REVERIE benchmark with a single stream of learning.

This paper presents a novel imitation learning framework, called constrained diffusion policy (CDP). The primary objective of CDP is to ensure that learned policies strictly adhere to safety constraints while imitating expert demonstrations. To achieve this, we define a polytopic constraint that represents the safe boundary for obstacle-free region. We introduce a novel mirror map and its inverse function to incorporate a generalized polytopic constraint manifold into the mirror diffusion model. By mapping sampled data onto a constrained manifold, the mirror diffusion model generates actions that satisfy safety constraints. This approach successfully addresses the safety issues commonly encountered in conventional imitation learning models. We apply the proposed framework to mobile navigation tasks in robotics, using the Isaac Gym simulator and the Unitree Go2 quadrupedal robot. Experimental results demonstrate that the proposed framework can successfully train policies that imitate expert behaviors while strictly maintaining safety constraints, thereby achieving safety-assured imitation learning.

Recent LiDAR-based 3D Object Detection (3DOD) methods show promising results, but they often do not generalize well to target domains outside the source (or training) data distribution. To reduce such domain gaps and thus to make 3DOD models more generalizable, we introduce a novel unsupervised domain adaptation (UDA) method, called CMDA, which (i) leverages visual semantic cues from an image modality (i.e., camera images) as an effective semantic bridge to close the domain gap in the cross-modal Bird's Eye View (BEV) representations. Further, (ii) we also introduce a self-training-based learning strategy, wherein a model is adversarially trained to generate domain-invariant features, which disrupt the discrimination of whether a feature instance comes from a source or an unseen target domain. Overall, our CMDA framework guides the 3DOD model to generate highly informative and domain-adaptive features for novel data distributions. In our extensive experiments with large-scale benchmarks, such as nuScenes, Waymo, and KITTI, those mentioned above provide significant performance gains for UDA tasks, achieving state-of-the-art performance.

Recent advances in 3D object detection leveraging multi-view cameras have demonstrated their practical and economical value in various challenging vision tasks. However, typical supervised learning approaches face challenges in achieving satisfactory adaptation toward unseen and unlabeled target datasets (i.e., direct transfer) due to the inevitable geometric misalignment between the source and target domains. In practice, we also encounter constraints on resources for training models and collecting annotations for the successful deployment of 3D object detectors. In this paper, we propose Unified Domain Generalization and Adaptation (UDGA), a practical solution to mitigate those drawbacks. We first propose Multi-view Overlap Depth Constraint that leverages the strong association between multi-view, significantly alleviating geometric gaps due to perspective view changes. Then, we present a Label-Efficient Domain Adaptation approach to handle unfamiliar targets with significantly fewer amounts of labels (i.e., 1$\%$ and 5$\%)$, while preserving well-defined source knowledge for training efficiency. Overall, UDGA framework enables stable detection performance in both source and target domains, effectively bridging inevitable domain gaps, while demanding fewer annotations. We demonstrate the robustness of UDGA with large-scale benchmarks: nuScenes, Lyft, and Waymo, where our framework outperforms the current state-of-the-art methods.

Predicting and constructing road geometric information (e.g., lane lines, road markers) is a crucial task for safe autonomous driving, while such static map elements can be repeatedly occluded by various dynamic objects on the road. Recent studies have shown significantly improved vectorized high-definition (HD) map construction performance, but there has been insufficient investigation of temporal information across adjacent input frames (i.e., clips), which may lead to inconsistent and suboptimal prediction results. To tackle this, we introduce a novel paradigm of clip-level vectorized HD map construction, MapUnveiler, which explicitly unveils the occluded map elements within a clip input by relating dense image representations with efficient clip tokens. Additionally, MapUnveiler associates inter-clip information through clip token propagation, effectively utilizing long- term temporal map information. MapUnveiler runs efficiently with the proposed clip-level pipeline by avoiding redundant computation with temporal stride while building a global map relationship. Our extensive experiments demonstrate that MapUnveiler achieves state-of-the-art performance on both the nuScenes and Argoverse2 benchmark datasets. We also showcase that MapUnveiler significantly outperforms state-of-the-art approaches in a challenging setting, achieving +10.7% mAP improvement in heavily occluded driving road scenes. The project page can be found at https://mapunveiler.github.io.

The estimation of 3D lanes from monocular RGB images is a fundamentally ill-posed problem. Previous studies have assumed that all lanes are on a flat ground plane. However, we argue that the algorithms based on this assumption have difficulty in detecting various lanes in actual driving environments. Contrary to previous approaches, we expand rich contextual features from an image domain to a 3D space by utilizing depth-aware voxel mapping. In addition, we determine 3D lanes based on voxelized features. We design a new lane representation combined with uncertainties and predict the confidence intervals of 3D lane points using Laplace loss. Experimental results show that the proposed method achieves state-of-the-art detection accuracy on three challenging datasets, including two real-world datasets, and significantly outperforms existing methods with reasonable computation load.

3D object detection (3DOD) from multi-view images is an economically appealing alternative to expensive LiDAR-based detectors, but also an extremely challenging task due to the absence of precise spatial cues. Recent studies have leveraged the teacher-student paradigm for cross-modal distillation, where a strong LiDAR-modality teacher transfers useful knowledge to a multi-view-based image-modality student. However, prior approaches have only focused on minimizing global distances between cross-modal features, which may lead to suboptimal knowledge distillation results. Based on these insights, we propose a novel structural and temporal cross-modal knowledge distillation (STXD) framework for multi-view 3DOD. First, STXD reduces redundancy of the feature components of the student by regularizing the cross-correlation of cross-modal features, while maximizing their similarities. Second, to effectively transfer temporal knowledge, STXD encodes temporal relations of features across a sequence of frames via similarity maps. Lastly, STXD also adopts a response distillation method to further enhance the quality of knowledge distillation at the output-level. Our extensive experiments demonstrate that STXD significantly improves the NDS and mAP of the based student detectors by 2.8%~4.5% on the nuScenes testing dataset.

In this paper, we study a practical omnidirectional depth estimation with neural networks that enables effective learning on real world data obtained using wide-baseline multiple fish-eye cameras. Most previous approaches only used synthetic data providing dense and accurate depth ground truth (GT). However, it is unrealistic to acquire such high quality GT data in real world due to limitations of the existing depth sensors. We first introduce two critical problems that can reduce the accuracy of depth estimation: depth GT sparsity and sensor calibration error. We then propose a novel semi-supervised learning method using pixel-level loss that selectively uses supervised loss and unsupervised re-projection loss according to existence of GT. Empirical results demonstrate that our method efficiently reduces the performance degradation in both simulation on synthetic data and real world data using sparse depth sensor.

We propose a fast and lightweight end-to-end convolutional network architecture for real-time segmentation of high resolution videos, NfS-SegNet, that can segement 2K-videos at 36.5 FPS with 24.3 GFLOPS. This speed and computation-efficiency is due to following reasons: 1) The encoder network, NfS-Net, is optimized for speed with simple building blocks without memory-heavy operations such as depthwise convolutions, and outperforms state-of-the-art lightweight CNN architectures such as SqueezeNet [2], Mo- bileNet v1 [3] & v2 [4] and ShuffleNet v1 [5] & v2 [6] on image classification with significantly higher speed. 2) The NfS- SegNet has an asymmetric architecture with deeper encoder and shallow decoder, whose design is based on our empirical finding that the decoder is the main bottleneck in computation with relatively small contribution to the final performance. 3) Our novel uncertainty-aware knowledge distillation method guides the teacher model to focus its knowledge transfer on the most difficult image regions. We validate the performance of NfS-SegNet with the CITYSCAPE [1] benchmark, on which it achieves state-of-the-art performance among lightweight segementation models in terms of both accuracy and speed.