WACV 2026 Papers

Single LDR to HDR reconstruction remains challenging for over-exposed regions where traditional methods often fail due to complete information loss. We present a training-free approach that enhances existing indirect HDR reconstruction methods through diffusion-based inpainting. Our method combines text-guided diffusion models with SDEdit refinement to generate plausible content in over-exposed areas while maintaining consistency across multi-exposure LDR images. Unlike previous approaches requiring extensive training, our method seamlessly integrates with existing indirect HDR reconstruction techniques through an iterative compensation mechanism that ensures luminance coherence across multiple exposures. We demonstrate significant improvements in both perceptual quality and quantitative metrics on standard HDR datasets and in-the-wild captures. Results show that our method effectively recovers natural details in challenging scenarios while preserving the advantages of existing HDR reconstruction pipelines.

Audio-driven talking face generation has received growing interest, particularly for applications requiring expressive and natural human-avatar interaction. However, most existing emotion-aware methods rely on a single modality (either audio or image) for emotion embedding, limiting their ability to capture nuanced affective cues. Additionally, most methods condition on a single reference image, restricting the model's ability to represent dynamic changes in actions or attributes across time. To address these issues, we introduce SynchroRaMa, a novel framework that integrates a multi-modal emotion embedding by combining emotional signals from text (via sentiment analysis) and audio (via speech-based emotion recognition and audio-derived valence-arousal features), enabling the generation of talking face videos with richer and more authentic emotional expressiveness and fidelity. To ensure natural head motion and accurate lip synchronization, SynchroRaMa includes an audio-to-motion (A2M) module that generates motion frames aligned with the input audio. Finally, SynchroRaMa incorporates scene descriptions generated by Large Language Model (LLM) as additional textual input, enabling it to capture dynamic actions and high-level semantic attributes. Conditioning the model on both visual and textual cues enhances temporal consistency and visual realism. Quantitative and qualitative experiments on benchmark datasets demonstrate that SynchroRaMa outperforms the state-of-the-art, achieving improvements in image quality, expression preservation, and motion realism. A user study further confirms that SynchroRaMa achieves higher subjective ratings than competing methods in overall naturalness, motion diversity, and video smoothness.

Object segmentation relies heavily on costly pixel-level annotations and struggles to generalize to unseen domains. The recent introduction of the Segment Anything Model (SAM), a foundation model for segmentation, offers a prompt-driven, zero-shot capability that has been applied in various domains (e.g., autonomous driving, satellite imagery, medical imaging) and extended to Few-Shot Segmentation (FSS) tasks. However, existing SAM-based FSS methods typically generate prompts by using a vision encoder to measure support-query image similarity, which often biases towards the support images and fails when there are significant support-query context shifts. To address this limitation, we propose a training-free FSS approach that combines visual and textual cues to generate effective prompts for the target class. By leveraging both vision and language information, our approach bridges the support-query gap and guides SAM to segment novel objects more reliably. Without any additional training, our method outperforms previous state-of-the-art FSS methods on established benchmarks (COCO\text - 20^i, Pascal\text - 5^i), demonstrating its effectiveness and robust generalization. Our code is publicly available on GitHub.

Effective image-based analysis of animals, their phenotypes, and their behavior requires accurate localization of key bodyparts. Keypoint detection algorithms can be either generalized, for a large set of animal species, or specialized and targeted to a specific small set of species. In this paper, we explore specialized models, and address two critical aspects of the associated data-label quality: selection of training data and definition of keypoints. Using antelope species as an example, we introduce a variety of species-similarity measures that we apply for selecting relevant training samples, and we demonstrate that training with the automatically selected species leads to improved pose estimation performance while reducing the required number of images. Then, we demonstrate that labeling keypoints with more precise locations leads to improved localization performance that would be valuable for downstream tasks.

Image-based localization in GNSS-denied environments is critical for UAV autonomy. State-of-the-art methods typically achieve this by matching onboard aerial images to a database of geo-referenced satellite images. However, these methods are fundamentally data-hungry, requiring large-scale, paired satellite and UAV imagery for training, which is often expensive and impractical. To address this, we propose a novel training paradigm that eliminates the need for any UAV data during training by learning to localize from satellite-view reference images alone. This is enabled by a data augmentation strategy that simulates the challenging visual domain shift from satellite to real-world UAV imagery. We introduce CAEVL, an efficient model designed to leverage this paradigm. To validate our approach, we release ViLD, a new, challenging dataset of real-world UAV images.

Human creativity follows a perceptual process, moving from abstract ideas to finer details during creation. While 3D generative models have advanced dramatically, models specifically designed to assist human imagination in 3D creation--particularly for detailing abstractions from coarse to fine--have not been explored. We propose a framework that enables intuitive and interactive 3D shape generation by iteratively splitting bounding boxes to refine the set of bounding boxes. The main technical components of our framework are two generative models: the box-splitting generative model and the box-to-shape generative model. The first model, named \Ours , generates a collection of 3D part bounding boxes with varying granularity by iteratively splitting coarse bounding boxes. It utilizes part bounding boxes created through agglomerative merging and learns the reverse of the merging process--the splitting sequences. The model consists of two main components: the first learns the categorical distribution of the box to be split, and the second learns the distribution of the two new boxes, given the set of boxes and the indication of which box to split. The second model, the box-to-shape generative model, is trained by leveraging the 3D shape priors learned by an existing 3D diffusion model while adapting the model to incorporate bounding box conditioning. In our experiments, we demonstrate that the box-splitting generative model outperforms token prediction models and the inpainting approach with an unconditional diffusion model. Also, we show that our box-to-shape model, based on a state-of-the-art 3D diffusion model, provides superior results compared to a previous model.

Vision-Language Models (VLMs) demonstrate remarkable capabilities in visual understanding and reasoning, such as in Visual Question Answering (VQA), where the model is asked a question related to a visual input. Still, these models can make distinctly unnatural errors, for example, providing (wrong) answers to unanswerable VQA questions, such as questions asking about objects that do not appear in the image. To address this issue, we propose CLIP-UP: CLIP-based Unanswerable Problem detection, a novel lightweight method for equipping VLMs with the ability to withhold answers to unanswerable questions. CLIP-UP leverages CLIP-based similarity measures to extract question-image alignment information to detect unanswerability, requiring efficient training of only a few additional layers, while keeping the original VLMs' weights unchanged. Tested across several models, CLIP-UP achieves significant improvements on benchmarks assessing unanswerability in both multiple-choice and open-ended VQA, surpassing other methods, while preserving original performance on other tasks.

High dynamic range (HDR) imaging aims to simultaneously capture a large range of illuminance levels, often observed in real world scenes. An HDR image is conventionally captured on a CMOS sensor by combining multiple frames with different exposure values. Ultimately, the dynamic range is limited by the read noise of the CMOS at low-light and by the full-well capacity at high-light. Single-photon avalanche diode (SPAD) imagers display high HDR capabilities due to sensitivity from individual photons up to high photon fluxes. In this work, we demonstrate that fusing multi-frame raw-CMOS RGB and monochrome SPAD images can enhance the dynamic range and color accuracy in real-world settings. To fully leverage the SPAD data, we propose a Linearisation-Upsample processing block that linearises the intrinsically nonlinear response of SPAD-QIS and accounts for the low spatial resolution of current commercial SPAD technology. We demonstrate a clear advantage by including a SPAD image into a multiframe HDR pipeline with comprehensive evaluation on HDR datasets and real-world data. We also address the lack of public SPAD datasets by providing two raw CMOS-SPAD datasets for multi-frame HDR. Both datasets are available for download here: https://github.com/suonsivu/raw-cmos-spad-hdr

Automating garment assembly from sewing patterns remains a significant challenge due to the lack of standardized annotation protocols and the frequent absence of semantic cues. Existing methods often rely on panel labels or handcrafted heuristics, which limit their applicability to real-world, non-conforming patterns. We present AutoSew, a fully automatic, geometry-based approach for predicting stitch correspondences directly from 2D pattern contours. AutoSew formulates the problem as a graph matching task, leveraging a Graph Neural Network to capture local and global geometric context, and employing a differentiable optimal transport solver to infer stitching relationships--including multi-edge connections. To support this task, we update the GarmentCodeData dataset modifying over 18k patterns with realistic multi-edge annotations, reflecting industrial assembly scenarios. AutoSew achieves 96% F1-score and successfully assembles 73.3% of test garments without error, outperforming existing methods while relying solely on geometric input. Our results demonstrate that geometry alone can robustly guide stitching prediction, enabling scalable garment assembly without manual input. Our dataset and code are available online.

Classical approaches to the rectification of a single image of a product, without stereo correspondences, require spatial landmarks. These landmarks, constructed from high-contrast elementary shapes that can be detected with simple algorithms, are highly conspicuous. To rectify complex deformations, one can use chessboard patterns of markers with elements that break quadrilateral symmetry, such as the three eyes of a QR code. However, these marker boards are even more conspicuous than a single marker. In traceability applications, only one site of marking is used, limiting the complexity of the surface on which it can be read, and exposing the mark to deidentification attacks for diversion of the product to a grey market. We introduce a method for constructing stealth and robust fiducial markers that can be displayed across a surface, limiting exposure to marker tampering for product deidentification. These markers, which we refer to as self-rectifying textures, can be used to rectify complex deformations by solving an inverse problem rather than relying on pixel correspondences of conspicuous landmarks. These stealth textures place fiducial markers in the autocorrelation of the image. In this way, crops of the deformed texture can be rectified using only these spatially invariant statistical properties. Affine transformations of an image correspond to linear transformations of the autocorrelation, without phase component. Exploiting this fact, self-rectifying textures enable the local estimation of the differential of a planar deformation by identifying landmarks in the autocorrelation image, such as peaks, whose locations in the fronto-parallel view of the texture are known. The translation component can be recovered independently via phase correlation. A rectifying map, modulo translations, can also be fit directly to local observations of the differential of the deformation, without access to the rectified texture or need for phase correlation. Self-rectifying textures can be used for communication, watermarking, authentication, surface identification, calibration, and geometry processing.

When spurious correlations between targets and data are present in training samples, deep neural networks may struggle to generalize, typically learning shortcuts corresponding to such undesired correlations (i.e., the bias), rather than fundamental target attributes. If bias attributes are assumed to be known, several strategies can be applied to mitigate a model's dependency on bias, including upsampling or upweighting samples with no bias. However, this is hardly the case in real-world scenarios, and alternative unsupervised debiasing approaches have been proposed in recent years, assuming no bias information is available. In this work, we propose Lose Your Self (LoYS), a novel bias-unsupervised approach for model debiasing. The main design strategy in LoYS aims to force a model learning semantic features, discouraging it from learning bias-related descriptors, specifically focusing on the target classifier confidence. Exploiting an adversarial scheme based on entropy loss, against a shallow auxiliary classifier trained to match the predictions of a pre-trained biased model, and entropy regularizations, our experiments show how LoYS is competitive or outperforms state-of-the-art methods, on several common benchmarks.

Open-vocabulary object detectors such as Grounding DINO are trained on vast and diverse data, achieving remarkable performance on challenging datasets. Due to that, it is unclear where to find their limitations, which is of major concern when using in safety-critical applications. Real-world data does not provide sufficient control, required for a rigorous evaluation of model generalization. In contrast, synthetically generated data allows to systematically explore the boundaries of model competence/generalization. In this work, we address two research questions: 1) Can we challenge open-vocabulary object detectors with generated image content? 2) Can we find systematic failure modes of those models? To address these questions, we design two automated pipelines using stable diffusion to inpaint unusual objects with high diversity in semantics, by sampling multiple substantives from WordNet and ChatGPT. On the synthetically generated data, we evaluate and compare multiple open-vocabulary object detectors as well as a classical object detector. The synthetic data is derived from two real-world datasets, namely LostAndFound, a challenging out-of-distribution (OOD) detection benchmark, and the NuImages dataset. Our results indicate that inpainting can challenge open-vocabulary object detectors in terms of overlooking objects. Additionally, we find a strong dependence of open-vocabulary models on object location, rather than on object semantics. This provides a systematic approach to challenge open-vocabulary models and gives valuable insights on how data could be acquired to effectively improve these models.

Semantic segmentation in hyperbolic space can capture hierarchical structure in low dimensions with uncertainty quantification. Existing approaches choose the Poincare ball model for hyperbolic geometry, which suffers from numerical instabilities, optimization, and computational challenges. We propose a novel, tractable, architecture-agnostic semantic segmentation framework in the hyperbolic Lorentz model. We employ text embeddings with semantic and visual cues to guide hierarchical pixel-level representations in Lorentz space. This enables stable and efficient optimization without requiring a Riemannian optimizer, and easily integrates with existing Euclidean architectures. Beyond segmentation, our approach yields free uncertainty estimation, confidence map, boundary delineation, hierarchical and text-based retrieval, and zero-shot performance, reaching generalized flatter minima. We further introduce a novel uncertainty and confidence indicator in Lorentz cone embeddings. Extensive experiments on ADE20K, COCO-Stuff-164k, Pascal-VOC, and Cityscapes with state-of-the-art models (DeepLabV3 and SegFormer) validate the effectiveness and generality of our approach. Our results demonstrate the potential of hyperbolic Lorentz embeddings for robust and uncertainty-aware semantic segmentation. Code is available at https://github.com/mxahan/Lorentz_semantic_segmentation.

Accurate dose calculation is critical in Gamma Knife Radiosurgery (GKRS), especially in regions near the skull where tissue heterogeneity can significantly alter dose distributions. Although the convolution algorithm-based dose calculation (Conv dose) using CT enables heterogeneity correction for an accurate treatment plan, it introduces additional clinical burdens, including longer planning times and increased radiation exposure, and routine acquisition of CT is not standard practice in GKRS. Moreover, Conv dose is computationally intensive, requiring up to several minutes for large targets, and in a clinical workflow where beam parameters and shot placements are iteratively adjusted, such delays can accumulate. Consequently, there is a growing demand for methods that can generate heterogeneity-aware convolution-equivalent doses in real time, on the order of a few seconds, to support smooth interactive plan adaptation. This study is the first to explore using the conditional wavelet Denoising Diffusion Probabilistic Model (cwDDPM) to generate radiation dose distributions. cwDDPM exploits the inherent sparsity of radiation dose maps in the wavelet domain to achieve more efficient learning and sampling. As a result, it produces synthetic Conv doses quickly and accurately, without relying on CT imaging or computationally intensive convolution calculations. Quantitative results across isodose overlap and dose-volume metrics demonstrate that cwDDPM achieves high fidelity to the ground truth Conv dose and performs comparably to state-of-the-art diffusion models, while reducing inference time by up to 45-fold. cwDDPM also demonstrates robustness across different tumor locations, especially near the skull, where heterogeneity correction is most critical. These results suggest that cwDDPM is a promising solution for rapid, CT-free Conv dose generation in GKRS planning.

2D-to-3D human pose lifting is an ill-posed problem due to depth ambiguity and occlusion. Existing methods relying on spatial and temporal consistency alone are insufficient to resolve these problems especially in the presence of significant occlusions or high dynamic actions. Semantic information, however, offers a complementary signal that can help disambiguate such cases. To this end, we propose LangPose, a framework that leverages action knowledge by aligning motion embeddings with text embeddings of fine-grained action labels. LangPose operates in two stages: pretraining and fine-tuning. In the pretraining stage, the model simultaneously learns to recognize actions and reconstruct 3D poses from masked and noisy 2D poses. During the fine-tuning stage, the model is further refined using real-world 3D human pose estimation datasets without action labels. Additionally, our framework incorporates masked body parts and masked time windows in motion modeling, encouraging the model to leverage semantic information when spatial and temporal consistency is unreliable. Experiments demonstrate the effectiveness of LangPose, achieving SOTA level performance in 3D pose estimation on public datasets, including Human3.6M and MPI-INF-3DHP. Specifically, LangPose achieves an MPJPE of 36.7mm on Human3.6M with detected 2D poses as input and 15.5mm on MPI-INF-3DHP with ground-truth 2D poses as input.

Change detection (CD) is critical in computer vision and remote sensing, with applications in monitoring, disaster response, and urban analysis. Most CD models assume co-registered inputs, but real imagery often suffers from parallax, viewpoint shifts, or long temporal gaps, leading to severe misalignment. Conventional register-then-detect pipelines and recent joint frameworks (e.g., BiFA, ChangeRD) remain limited: they rely on regression-only flow, global homographies, or synthetic perturbations that fail under large displacements. We propose DiffRegCD, an integrated framework that couples dense registration and change detection. DiffRegCD reformulates correspondence as a Gaussian-smoothed classification task, delivering sub-pixel accuracy and stable training. It builds on frozen multi-scale features from a pretrained denoising diffusion model, which provide invariance to viewpoint and illumination variation. Supervision is enabled by controlled affine perturbations applied to standard CD datasets, yielding paired ground truth for both flow and change detection without pseudo-labels. Experiments on aerial (LEVIR-CD, DSIFN-CD, WHU-CD, SYSU-CD) and ground-level (VL-CMU-CD) datasets show that DiffRegCD outperforms recent baselines and remains robust under wide temporal and viewpoint variation, establishing diffusion features and classification-based correspondence as a strong foundation for integrated CD.

Event-based cameras are well-suited for fast and agile autonomous navigation due to their ultra-fast, microsecond level temporal resolution. However, fully leveraging this potential requires highly efficient processing algorithms capable of asynchronous, event-by-event representations and policy updates. Current methods employ synchronous dense representation or process events in a fixed-rate time windows, leading to inefficiencies via redundant computation. We address this by proposing an end-to-end framework for event-to-control policy learning designed for reactive navigation tasks. Our method consists of a memory-augmented perception module that updates the representation asynchronously and adaptively selects the number of events to process. Using the memory representation, a lightweight policy module is jointly optimized with the perception module, and learns to predict control commands at rates that dynamically adjust to scene complexity in an event-based reinforcement learning setting. Evaluations on simulated drone navigation tasks demonstrate higher sample efficiency and robustness compared to dense frame-based methods. Moreover, our approach significantly reduces computational complexity by minimizing processing steps and event counts while maintaining competitive performance against state-of-the-art event-based methods.

The costly annotation process has driven the development of semi-supervised learning (SSL) approaches. Existing semi-supervised LiDAR segmentation methods typically process entire point clouds directly, aiming to assign labels to all points at the scene scale. However, the large number of points, combined with their sparse and irregular nature, makes it challenging to learn scene-level optimization objectives, especially in SSL settings where labeled data are insufficient. This paper presents a Cluster-based pseudo-LAbeling Semi-Supervised technique, called CLASS. CLASS is designed to divide point clouds into several small, pure clusters, thereby decomposing challenging scene-scale segmentation task into more manageable cluster-scale classification and segmentation tasks, enabling the generation of high-quality pseudo labels for unlabeled data. CLASS possesses three key properties. i) Task simplicity: our pseudo-labeling process is based on simpler cluster-scale classification and segmentation tasks, resulting in ease of learning. ii) Labeling effectiveness: CLASS can generate pseudo-labels comparable to ground truth using only approximately 10% labeled data. iii) Universal versatility: CLASS exhibits flexibility regarding LiDAR representations (e.g., BEV, voxel, and range view). Comprehensive experiments on popular LiDAR segmentation benchmarks demonstrate its superiority.

3D Generative Models that synthesize high fidelity 3D assets from single view or multi-view images cannot recover precise geometry and real-world object measurements needed for practical applications. On the other hand, multi-view 3D reconstruction methods based on structure from motion, implicit surfaces, gaussian splatting, fail to recover high fidelity object meshes with dense geometry, shape regularity, smooth surfaces present in real world objects. In this paper we propose a novel approach that leverages a 3D mesh prior synthesized by generative models pre-trained on large scale 3D synthetic datasets. Our method refines the initial mesh geometry without use of any additional training data while improving accuracy of 3D geometry via multi-view consistency, without degrading the mesh surface quality. Our method can automatically reconstruct meshes from images of objects in real world scenes, without requiring any additional large-scale training data or manual inputs.