Victor Adrian Prisacariu

Wenjing Bian, Zirui Wang, Kejie Li et al. · bytedance, oxford

Training a Neural Radiance Field (NeRF) without pre-computed camera poses is challenging. Recent advances in this direction demonstrate the possibility of jointly optimising a NeRF and camera poses in forward-facing scenes. However, these methods still face difficulties during dramatic camera movement. We tackle this challenging problem by incorporating undistorted monocular depth priors. These priors are generated by correcting scale and shift parameters during training, with which we are then able to constrain the relative poses between consecutive frames. This constraint is achieved using our proposed novel loss functions. Experiments on real-world indoor and outdoor scenes show that our method can handle challenging camera trajectories and outperforms existing methods in terms of novel view rendering quality and pose estimation accuracy. Our project page is https://nope-nerf.active.vision.

Kejie Li, Jia-Wang Bian, Robert Castle et al. · bytedance, oxford

High-quality 3D ground-truth shapes are critical for 3D object reconstruction evaluation. However, it is difficult to create a replica of an object in reality, and even 3D reconstructions generated by 3D scanners have artefacts that cause biases in evaluation. To address this issue, we introduce a novel multi-view RGBD dataset captured using a mobile device, which includes highly precise 3D ground-truth annotations for 153 object models featuring a diverse set of 3D structures. We obtain precise 3D ground-truth shape without relying on high-end 3D scanners by utilising LEGO models with known geometry as the 3D structures for image capture. The distinct data modality offered by high-resolution RGB images and low-resolution depth maps captured on a mobile device, when combined with precise 3D geometry annotations, presents a unique opportunity for future research on high-fidelity 3D reconstruction. Furthermore, we evaluate a range of 3D reconstruction algorithms on the proposed dataset. Project page: http://code.active.vision/MobileBrick/

Shuai Chen, Xinghui Li, Zirui Wang et al. · oxford

We introduce a camera relocalization pipeline that combines absolute pose regression (APR) and direct feature matching. By incorporating exposure-adaptive novel view synthesis, our method successfully addresses photometric distortions in outdoor environments that existing photometric-based methods fail to handle. With domain-invariant feature matching, our solution improves pose regression accuracy using semi-supervised learning on unlabeled data. In particular, the pipeline consists of two components: Novel View Synthesizer and DFNet. The former synthesizes novel views compensating for changes in exposure and the latter regresses camera poses and extracts robust features that close the domain gap between real images and synthetic ones. Furthermore, we introduce an online synthetic data generation scheme. We show that these approaches effectively enhance camera pose estimation both in indoor and outdoor scenes. Hence, our method achieves a state-of-the-art accuracy by outperforming existing single-image APR methods by as much as 56%, comparable to 3D structure-based methods.

Shuai Chen, Yash Bhalgat, Xinghui Li et al. · bytedance, oxford

Absolute Pose Regression (APR) methods use deep neural networks to directly regress camera poses from RGB images. However, the predominant APR architectures only rely on 2D operations during inference, resulting in limited accuracy of pose estimation due to the lack of 3D geometry constraints or priors. In this work, we propose a test-time refinement pipeline that leverages implicit geometric constraints using a robust feature field to enhance the ability of APR methods to use 3D information during inference. We also introduce a novel Neural Feature Synthesizer (NeFeS) model, which encodes 3D geometric features during training and directly renders dense novel view features at test time to refine APR methods. To enhance the robustness of our model, we introduce a feature fusion module and a progressive training strategy. Our proposed method achieves state-of-the-art single-image APR accuracy on indoor and outdoor datasets.

Kejie Li, Yansong Tang, Victor Adrian Prisacariu et al.

Dense 3D reconstruction from a stream of depth images is the key to many mixed reality and robotic applications. Although methods based on Truncated Signed Distance Function (TSDF) Fusion have advanced the field over the years, the TSDF volume representation is confronted with striking a balance between the robustness to noisy measurements and maintaining the level of detail. We present Bi-level Neural Volume Fusion (BNV-Fusion), which leverages recent advances in neural implicit representations and neural rendering for dense 3D reconstruction. In order to incrementally integrate new depth maps into a global neural implicit representation, we propose a novel bi-level fusion strategy that considers both efficiency and reconstruction quality by design. We evaluate the proposed method on multiple datasets quantitatively and qualitatively, demonstrating a significant improvement over existing methods.

Jia-Wang Bian, Wenjing Bian, Victor Adrian Prisacariu et al. · bytedance, oxford

Neural surface reconstruction is sensitive to the camera pose noise, even if state-of-the-art pose estimators like COLMAP or ARKit are used. More importantly, existing Pose-NeRF joint optimisation methods have struggled to improve pose accuracy in challenging real-world scenarios. To overcome the challenges, we introduce the pose residual field (PoRF), a novel implicit representation that uses an MLP for regressing pose updates. This is more robust than the conventional pose parameter optimisation due to parameter sharing that leverages global information over the entire sequence. Furthermore, we propose an epipolar geometry loss to enhance the supervision that leverages the correspondences exported from COLMAP results without the extra computational overhead. Our method yields promising results. On the DTU dataset, we reduce the rotation error by 78\% for COLMAP poses, leading to the decreased reconstruction Chamfer distance from 3.48mm to 0.85mm. On the MobileBrick dataset that contains casually captured unbounded 360-degree videos, our method refines ARKit poses and improves the reconstruction F1 score from 69.18 to 75.67, outperforming that with the dataset provided ground-truth pose (75.14). These achievements demonstrate the efficacy of our approach in refining camera poses and improving the accuracy of neural surface reconstruction in real-world scenarios.

Eduardo Arnold, Jamie Wynn, Sara Vicente et al.

Can we relocalize in a scene represented by a single reference image? Standard visual relocalization requires hundreds of images and scale calibration to build a scene-specific 3D map. In contrast, we propose Map-free Relocalization, i.e., using only one photo of a scene to enable instant, metric scaled relocalization. Existing datasets are not suitable to benchmark map-free relocalization, due to their focus on large scenes or their limited variability. Thus, we have constructed a new dataset of 655 small places of interest, such as sculptures, murals and fountains, collected worldwide. Each place comes with a reference image to serve as a relocalization anchor, and dozens of query images with known, metric camera poses. The dataset features changing conditions, stark viewpoint changes, high variability across places, and queries with low to no visual overlap with the reference image. We identify two viable families of existing methods to provide baseline results: relative pose regression, and feature matching combined with single-image depth prediction. While these methods show reasonable performance on some favorable scenes in our dataset, map-free relocalization proves to be a challenge that requires new, innovative solutions.

Brandon Smart, Chuanxia Zheng, Iro Laina et al.

In this paper, we introduce Splatt3R, a pose-free, feed-forward method for in-the-wild 3D reconstruction and novel view synthesis from stereo pairs. Given uncalibrated natural images, Splatt3R can predict 3D Gaussian Splats without requiring any camera parameters or depth information. For generalizability, we build Splatt3R upon a ``foundation'' 3D geometry reconstruction method, MASt3R, by extending it to deal with both 3D structure and appearance. Specifically, unlike the original MASt3R which reconstructs only 3D point clouds, we predict the additional Gaussian attributes required to construct a Gaussian primitive for each point. Hence, unlike other novel view synthesis methods, Splatt3R is first trained by optimizing the 3D point cloud's geometry loss, and then a novel view synthesis objective. By doing this, we avoid the local minima present in training 3D Gaussian Splats from stereo views. We also propose a novel loss masking strategy that we empirically find is critical for strong performance on extrapolated viewpoints. We train Splatt3R on the ScanNet++ dataset and demonstrate excellent generalisation to uncalibrated, in-the-wild images. Splatt3R can reconstruct scenes at 4FPS at 512 x 512 resolution, and the resultant splats can be rendered in real-time.

Yiwen Li, Yunguan Fu, Iani Gayo et al.

The prowess that makes few-shot learning desirable in medical image analysis is the efficient use of the support image data, which are labelled to classify or segment new classes, a task that otherwise requires substantially more training images and expert annotations. This work describes a fully 3D prototypical few-shot segmentation algorithm, such that the trained networks can be effectively adapted to clinically interesting structures that are absent in training, using only a few labelled images from a different institute. First, to compensate for the widely recognised spatial variability between institutions in episodic adaptation of novel classes, a novel spatial registration mechanism is integrated into prototypical learning, consisting of a segmentation head and an spatial alignment module. Second, to assist the training with observed imperfect alignment, support mask conditioning module is proposed to further utilise the annotation available from the support images. Extensive experiments are presented in an application of segmenting eight anatomical structures important for interventional planning, using a data set of 589 pelvic T2-weighted MR images, acquired at seven institutes. The results demonstrate the efficacy in each of the 3D formulation, the spatial registration, and the support mask conditioning, all of which made positive contributions independently or collectively. Compared with the previously proposed 2D alternatives, the few-shot segmentation performance was improved with statistical significance, regardless whether the support data come from the same or different institutes.

Axel Barroso-Laguna, Eric Brachmann, Victor Adrian Prisacariu et al.

Camera pose estimation for two-view geometry traditionally relies on RANSAC. Normally, a multitude of image correspondences leads to a pool of proposed hypotheses, which are then scored to find a winning model. The inlier count is generally regarded as a reliable indicator of "consensus". We examine this scoring heuristic, and find that it favors disappointing models under certain circumstances. As a remedy, we propose the Fundamental Scoring Network (FSNet), which infers a score for a pair of overlapping images and any proposed fundamental matrix. It does not rely on sparse correspondences, but rather embodies a two-view geometry model through an epipolar attention mechanism that predicts the pose error of the two images. FSNet can be incorporated into traditional RANSAC loops. We evaluate FSNet on fundamental and essential matrix estimation on indoor and outdoor datasets, and establish that FSNet can successfully identify good poses for pairs of images with few or unreliable correspondences. Besides, we show that naively combining FSNet with MAGSAC++ scoring approach achieves state of the art results.

Theo W. Costain, Victor Adrian Prisacariu · oxford

We present ApproxConv, a novel method for compressing the layers of a convolutional neural network. Reframing conventional discrete convolution as continuous convolution of parametrised functions over space, we use functional approximations to capture the essential structures of CNN filters with fewer parameters than conventional operations. Our method is able to reduce the size of trained CNN layers requiring only a small amount of fine-tuning. We show that our method is able to compress existing deep network models by half whilst losing only 1.86% accuracy. Further, we demonstrate that our method is compatible with other compression methods like quantisation allowing for further reductions in model size.

Xianzheng Ma, Tao Sun, Shuai Chen et al.

Recent 3D Large-Language Models (3D-LLMs) claim to understand 3D worlds, especially spatial relationships among objects. Yet, we find that simply fine-tuning a language model on text-only question-answer pairs can perform comparably or even surpass these methods on the SQA3D benchmark without using any 3D input. This indicates that the SQA3D benchmark may not be able to detect if the model exploits textual shortcuts rather than engages in 3D-aware reasoning. To address this issue, we introduce Real-3DQA, a more rigorous evaluation benchmark that filters out easy-to-guess questions and introduces a structured taxonomy to assess various aspects of 3D reasoning. Experiments on Real-3DQA confirm that existing 3D-LLMs struggle with spatial relationships once simple cues are removed. We further propose a 3D-reweighted training objective that guides model to rely more on 3D visual clues, substantially enhancing 3D-LLMs performance in spatial reasoning tasks. Our findings underscore the need for robust benchmarks and tailored training strategies to advance genuine 3D vision-language understanding. Project page: https://real-3dqa.github.io/.

Changkun Liu, Shuai Chen, Yash Bhalgat et al.

We leverage 3D Gaussian Splatting (3DGS) as a scene representation and propose a novel test-time camera pose refinement (CPR) framework, GS-CPR. This framework enhances the localization accuracy of state-of-the-art absolute pose regression and scene coordinate regression methods. The 3DGS model renders high-quality synthetic images and depth maps to facilitate the establishment of 2D-3D correspondences. GS-CPR obviates the need for training feature extractors or descriptors by operating directly on RGB images, utilizing the 3D foundation model, MASt3R, for precise 2D matching. To improve the robustness of our model in challenging outdoor environments, we incorporate an exposure-adaptive module within the 3DGS framework. Consequently, GS-CPR enables efficient one-shot pose refinement given a single RGB query and a coarse initial pose estimation. Our proposed approach surpasses leading NeRF-based optimization methods in both accuracy and runtime across indoor and outdoor visual localization benchmarks, achieving new state-of-the-art accuracy on two indoor datasets. The project page is available at https://xrim-lab.github.io/GS-CPR/.

Shuai Chen, Tommaso Cavallari, Victor Adrian Prisacariu et al.

Pose regression networks predict the camera pose of a query image relative to a known environment. Within this family of methods, absolute pose regression (APR) has recently shown promising accuracy in the range of a few centimeters in position error. APR networks encode the scene geometry implicitly in their weights. To achieve high accuracy, they require vast amounts of training data that, realistically, can only be created using novel view synthesis in a days-long process. This process has to be repeated for each new scene again and again. We present a new approach to pose regression, map-relative pose regression (marepo), that satisfies the data hunger of the pose regression network in a scene-agnostic fashion. We condition the pose regressor on a scene-specific map representation such that its pose predictions are relative to the scene map. This allows us to train the pose regressor across hundreds of scenes to learn the generic relation between a scene-specific map representation and the camera pose. Our map-relative pose regressor can be applied to new map representations immediately or after mere minutes of fine-tuning for the highest accuracy. Our approach outperforms previous pose regression methods by far on two public datasets, indoor and outdoor. Code is available: https://nianticlabs.github.io/marepo

Xianzheng Ma, Brandon Smart, Yash Bhalgat et al. · bytedance, oxford

As large language models (LLMs) evolve, their integration with 3D spatial data (3D-LLMs) has seen rapid progress, offering unprecedented capabilities for understanding and interacting with physical spaces. This survey provides a comprehensive overview of the methodologies enabling LLMs to process, understand, and generate 3D data. Highlighting the unique advantages of LLMs, such as in-context learning, step-by-step reasoning, open-vocabulary capabilities, and extensive world knowledge, we underscore their potential to significantly advance spatial comprehension and interaction within embodied Artificial Intelligence (AI) systems. Our investigation spans various 3D data representations, from point clouds to Neural Radiance Fields (NeRFs). It examines their integration with LLMs for tasks such as 3D scene understanding, captioning, question-answering, and dialogue, as well as LLM-based agents for spatial reasoning, planning, and navigation. The paper also includes a brief review of other methods that integrate 3D and language. The meta-analysis presented in this paper reveals significant progress yet underscores the necessity for novel approaches to harness the full potential of 3D-LLMs. Hence, with this paper, we aim to chart a course for future research that explores and expands the capabilities of 3D-LLMs in understanding and interacting with the complex 3D world. To support this survey, we have established a project page where papers related to our topic are organized and listed: https://github.com/ActiveVisionLab/Awesome-LLM-3D.

Zirui Wang, Victor Adrian Prisacariu

We introduce a novel self-attention-based normal estimation network that is able to focus softly on relevant points and adjust the softness by learning a temperature parameter, making it able to work naturally and effectively within a large neighbourhood range. As a result, our model outperforms all existing normal estimation algorithms by a large margin, achieving 94.1% accuracy in comparison with the previous state of the art of 91.2%, with a 25x smaller model and 12x faster inference time. We also use point-to-plane Iterative Closest Point (ICP) as an application case to show that our normal estimations lead to faster convergence than normal estimations from other methods, without manually fine-tuning neighbourhood range parameters. Code available at https://code.active.vision.

Carl Yuheng Ren, Victor Adrian Prisacariu, Ian D Reid

We introduce a parallel GPU implementation of the Simple Linear Iterative Clustering (SLIC) superpixel segmentation. Using a single graphic card, our implementation achieves speedups of up to $83\times$ from the standard sequential implementation. Our implementation is fully compatible with the standard sequential implementation and the software is now available online and is open source.

Jing Wu, Jia-Wang Bian, Xinghui Li et al. · bytedance, oxford

We propose GaussCtrl, a text-driven method to edit a 3D scene reconstructed by the 3D Gaussian Splatting (3DGS). Our method first renders a collection of images by using the 3DGS and edits them by using a pre-trained 2D diffusion model (ControlNet) based on the input prompt, which is then used to optimise the 3D model. Our key contribution is multi-view consistent editing, which enables editing all images together instead of iteratively editing one image while updating the 3D model as in previous works. It leads to faster editing as well as higher visual quality. This is achieved by the two terms: (a) depth-conditioned editing that enforces geometric consistency across multi-view images by leveraging naturally consistent depth maps. (b) attention-based latent code alignment that unifies the appearance of edited images by conditioning their editing to several reference views through self and cross-view attention between images' latent representations. Experiments demonstrate that our method achieves faster editing and better visual results than previous state-of-the-art methods.

Eric Brachmann, Jamie Wynn, Shuai Chen et al.

We address the task of estimating camera parameters from a set of images depicting a scene. Popular feature-based structure-from-motion (SfM) tools solve this task by incremental reconstruction: they repeat triangulation of sparse 3D points and registration of more camera views to the sparse point cloud. We re-interpret incremental structure-from-motion as an iterated application and refinement of a visual relocalizer, that is, of a method that registers new views to the current state of the reconstruction. This perspective allows us to investigate alternative visual relocalizers that are not rooted in local feature matching. We show that scene coordinate regression, a learning-based relocalization approach, allows us to build implicit, neural scene representations from unposed images. Different from other learning-based reconstruction methods, we do not require pose priors nor sequential inputs, and we optimize efficiently over thousands of images. In many cases, our method, ACE0, estimates camera poses with an accuracy close to feature-based SfM, as demonstrated by novel view synthesis. Project page: https://nianticlabs.github.io/acezero/

Axel Barroso-Laguna, Sowmya Munukutla, Victor Adrian Prisacariu et al.

Given two images, we can estimate the relative camera pose between them by establishing image-to-image correspondences. Usually, correspondences are 2D-to-2D and the pose we estimate is defined only up to scale. Some applications, aiming at instant augmented reality anywhere, require scale-metric pose estimates, and hence, they rely on external depth estimators to recover the scale. We present MicKey, a keypoint matching pipeline that is able to predict metric correspondences in 3D camera space. By learning to match 3D coordinates across images, we are able to infer the metric relative pose without depth measurements. Depth measurements are also not required for training, nor are scene reconstructions or image overlap information. MicKey is supervised only by pairs of images and their relative poses. MicKey achieves state-of-the-art performance on the Map-Free Relocalisation benchmark while requiring less supervision than competing approaches.

Zirui Wang, Wenjing Bian, Victor Adrian Prisacariu

We introduce a novel cross-reference image quality assessment method that effectively fills the gap in the image assessment landscape, complementing the array of established evaluation schemes -- ranging from full-reference metrics like SSIM, no-reference metrics such as NIQE, to general-reference metrics including FID, and Multi-modal-reference metrics, e.g., CLIPScore. Utilising a neural network with the cross-attention mechanism and a unique data collection pipeline from NVS optimisation, our method enables accurate image quality assessment without requiring ground truth references. By comparing a query image against multiple views of the same scene, our method addresses the limitations of existing metrics in novel view synthesis (NVS) and similar tasks where direct reference images are unavailable. Experimental results show that our method is closely correlated to the full-reference metric SSIM, while not requiring ground truth references.

Zirui Wang, Wenjing Bian, Xinghui Li et al. · oxford

We introduce Oxford Day-and-Night, a large-scale, egocentric dataset for novel view synthesis (NVS) and visual relocalisation under challenging lighting conditions. Existing datasets often lack crucial combinations of features such as ground-truth 3D geometry, wide-ranging lighting variation, and full 6DoF motion. Oxford Day-and-Night addresses these gaps by leveraging Meta ARIA glasses to capture egocentric video and applying multi-session SLAM to estimate camera poses, reconstruct 3D point clouds, and align sequences captured under varying lighting conditions, including both day and night. The dataset spans over 30 $\mathrm{km}$ of recorded trajectories and covers an area of 40,000 $\mathrm{m}^2$, offering a rich foundation for egocentric 3D vision research. It supports two core benchmarks, NVS and relocalisation, providing a unique platform for evaluating models in realistic and diverse environments.

Wenjing Bian, Axel Barroso-Laguna, Tommaso Cavallari et al.

Scene coordinate regression (SCR) models have proven to be powerful implicit scene representations for 3D vision, enabling visual relocalization and structure-from-motion. SCR models are trained specifically for one scene. If training images imply insufficient multi-view constraints SCR models degenerate. We present a probabilistic reinterpretation of training SCR models, which allows us to infuse high-level reconstruction priors. We investigate multiple such priors, ranging from simple priors over the distribution of reconstructed depth values to learned priors over plausible scene coordinate configurations. For the latter, we train a 3D point cloud diffusion model on a large corpus of indoor scans. Our priors push predicted 3D scene points towards plausible geometry at each training step to increase their likelihood. On three indoor datasets our priors help learning better scene representations, resulting in more coherent scene point clouds, higher registration rates and better camera poses, with a positive effect on down-stream tasks such as novel view synthesis and camera relocalization.

Leonard Bruns, Axel Barroso-Laguna, Tommaso Cavallari et al.

Scene coordinate regression (SCR) has established itself as a promising learning-based approach to visual relocalization. After mere minutes of scene-specific training, SCR models estimate camera poses of query images with high accuracy. Still, SCR methods fall short of the generalization capabilities of more classical feature-matching approaches. When imaging conditions of query images, such as lighting or viewpoint, are too different from the training views, SCR models fail. Failing to generalize is an inherent limitation of previous SCR frameworks, since their training objective is to encode the training views in the weights of the coordinate regressor itself. The regressor essentially overfits to the training views, by design. We propose to separate the coordinate regressor and the map representation into a generic transformer and a scene-specific map code. This separation allows us to pre-train the transformer on tens of thousands of scenes. More importantly, it allows us to train the transformer to generalize from mapping images to unseen query images during pre-training. We demonstrate on multiple challenging relocalization datasets that our method, ACE-G, leads to significantly increased robustness while keeping the computational footprint attractive.

Axel Barroso-Laguna, Tommaso Cavallari, Victor Adrian Prisacariu et al.

Visually localizing an image, i.e., estimating its camera pose, requires building a scene representation that serves as a visual map. The representation we choose has direct consequences towards the practicability of our system. Even when starting from mapping images with known camera poses, state-of-the-art approaches still require hours of mapping time in the worst case, and several minutes in the best. This work raises the question whether we can achieve competitive accuracy much faster. We introduce FastForward, a method that creates a map representation and relocalizes a query image on-the-fly in a single feed-forward pass. At the core, we represent multiple mapping images as a collection of features anchored in 3D space. FastForward utilizes these mapping features to predict image-to-scene correspondences for the query image, enabling the estimation of its camera pose. We couple FastForward with image retrieval and achieve state-of-the-art accuracy when compared to other approaches with minimal map preparation time. Furthermore, FastForward demonstrates robust generalization to unseen domains, including challenging large-scale outdoor environments.

Jing Wu, Zirui Wang, Iro Laina et al.

Mirror reflections are common in everyday environments and can provide stereo information within a single capture, as the real and reflected virtual views are visible simultaneously. We exploit this property by treating the reflection as an auxiliary view and designing a transformation that constructs a physically valid virtual camera, allowing direct pixel-domain generation of the virtual view while adhering to the real-world imaging process. This enables a multi-view stereo setup from a single image, simplifying the imaging process, making it compatible with powerful feed-forward reconstruction models for generalizable and robust 3D reconstruction. To further exploit the geometric symmetry introduced by mirrors, we propose a symmetric-aware loss to refine pose estimation. Our framework also naturally extends to dynamic scenes, where each frame contains a mirror reflection, enabling efficient per-frame geometry recovery. For quantitative evaluation, we provide a fully customizable synthetic dataset of 16 Blender scenes, each with ground-truth point clouds and camera poses. Extensive experiments on real-world data and synthetic data are conducted to illustrate the effectiveness of our method.

Zirui Wang, Yash Bhalgat, Ruining Li et al.

We tackle active view selection in novel view synthesis and 3D reconstruction. Existing methods like FisheRF and ActiveNeRF select the next best view by minimizing uncertainty or maximizing information gain in 3D, but they require specialized designs for different 3D representations and involve complex modelling in 3D space. Instead, we reframe this as a 2D image quality assessment (IQA) task, selecting views where current renderings have the lowest quality. Since ground-truth images for candidate views are unavailable, full-reference metrics like PSNR and SSIM are inapplicable, while no-reference metrics, such as MUSIQ and MANIQA, lack the essential multi-view context. Inspired by a recent cross-referencing quality framework CrossScore, we train a model to predict SSIM within a multi-view setup and use it to guide view selection. Our cross-reference IQA framework achieves substantial quantitative and qualitative improvements across standard benchmarks, while being agnostic to 3D representations, and runs 14-33 times faster than previous methods.

Eric Brachmann, Tommaso Cavallari, Victor Adrian Prisacariu

Learning-based visual relocalizers exhibit leading pose accuracy, but require hours or days of training. Since training needs to happen on each new scene again, long training times make learning-based relocalization impractical for most applications, despite its promise of high accuracy. In this paper we show how such a system can actually achieve the same accuracy in less than 5 minutes. We start from the obvious: a relocalization network can be split in a scene-agnostic feature backbone, and a scene-specific prediction head. Less obvious: using an MLP prediction head allows us to optimize across thousands of view points simultaneously in each single training iteration. This leads to stable and extremely fast convergence. Furthermore, we substitute effective but slow end-to-end training using a robust pose solver with a curriculum over a reprojection loss. Our approach does not require privileged knowledge, such a depth maps or a 3D model, for speedy training. Overall, our approach is up to 300x faster in mapping than state-of-the-art scene coordinate regression, while keeping accuracy on par.

Xinghui Li, Kai Han, Xingchen Wan et al.

We propose SimSC, a remarkably simple framework, to address the problem of semantic matching only based on the feature backbone. We discover that when fine-tuning ImageNet pre-trained backbone on the semantic matching task, L2 normalization of the feature map, a standard procedure in feature matching, produces an overly smooth matching distribution and significantly hinders the fine-tuning process. By setting an appropriate temperature to the softmax, this over-smoothness can be alleviated and the quality of features can be substantially improved. We employ a learning module to predict the optimal temperature for fine-tuning feature backbones. This module is trained together with the backbone and the temperature is updated online. We evaluate our method on three public datasets and demonstrate that we can achieve accuracy on par with state-of-the-art methods under the same backbone without using a learned matching head. Our method is versatile and works on various types of backbones. We show that the accuracy of our framework can be easily improved by coupling it with more powerful backbones.

Yiwen Li, Yunguan Fu, Qianye Yang et al.

The ability to adapt medical image segmentation networks for a novel class such as an unseen anatomical or pathological structure, when only a few labelled examples of this class are available from local healthcare providers, is sought-after. This potentially addresses two widely recognised limitations in deploying modern deep learning models to clinical practice, expertise-and-labour-intensive labelling and cross-institution generalisation. This work presents the first 3D few-shot interclass segmentation network for medical images, using a labelled multi-institution dataset from prostate cancer patients with eight regions of interest. We propose an image alignment module registering the predicted segmentation of both query and support data, in a standard prototypical learning algorithm, to a reference atlas space. The built-in registration mechanism can effectively utilise the prior knowledge of consistent anatomy between subjects, regardless whether they are from the same institution or not. Experimental results demonstrated that the proposed registration-assisted prototypical learning significantly improved segmentation accuracy (p-values<0.01) on query data from a holdout institution, with varying availability of support data from multiple institutions. We also report the additional benefits of the proposed 3D networks with 75% fewer parameters and an arguably simpler implementation, compared with existing 2D few-shot approaches that segment 2D slices of volumetric medical images.

Yiwen Li, Gratianus Wesley Putra Data, Yunguan Fu et al.

Despite the success of deep learning methods for semantic segmentation, few-shot semantic segmentation remains a challenging task due to the limited training data and the generalisation requirement for unseen classes. While recent progress has been particularly encouraging, we discover that existing methods tend to have poor performance in terms of meanIoU when query images contain other semantic classes besides the target class. To address this issue, we propose a novel self-supervised task that generates random pseudo-classes in the background of the query images, providing extra training data that would otherwise be unavailable when predicting individual target classes. To that end, we adopted superpixel segmentation for generating the pseudo-classes. With this extra supervision, we improved the meanIoU performance of the state-of-the-art method by 2.5% and 5.1% on the one-shot tasks, as well as 6.7% and 4.4% on the five-shot tasks, on the PASCAL-5i and COCO benchmarks, respectively.

Wenjing Bian, Zirui Wang, Kejie Li et al.

We propose Ray-ONet to reconstruct detailed 3D models from monocular images efficiently. By predicting a series of occupancy probabilities along a ray that is back-projected from a pixel in the camera coordinate, our method Ray-ONet improves the reconstruction accuracy in comparison with Occupancy Networks (ONet), while reducing the network inference complexity to O($N^2$). As a result, Ray-ONet achieves state-of-the-art performance on the ShapeNet benchmark with more than 20$\times$ speed-up at $128^3$ resolution and maintains a similar memory footprint during inference.

Henry Howard-Jenkins, Jose-Raul Ruiz-Sarmiento, Victor Adrian Prisacariu

We present LaLaLoc to localise in environments without the need for prior visitation, and in a manner that is robust to large changes in scene appearance, such as a full rearrangement of furniture. Specifically, LaLaLoc performs localisation through latent representations of room layout. LaLaLoc learns a rich embedding space shared between RGB panoramas and layouts inferred from a known floor plan that encodes the structural similarity between locations. Further, LaLaLoc introduces direct, cross-modal pose optimisation in its latent space. Thus, LaLaLoc enables fine-grained pose estimation in a scene without the need for prior visitation, as well as being robust to dynamics, such as a change in furniture configuration. We show that in a domestic environment LaLaLoc is able to accurately localise a single RGB panorama image to within 8.3cm, given only a floor plan as a prior.

Zirui Wang, Shangzhe Wu, Weidi Xie et al.

Considering the problem of novel view synthesis (NVS) from only a set of 2D images, we simplify the training process of Neural Radiance Field (NeRF) on forward-facing scenes by removing the requirement of known or pre-computed camera parameters, including both intrinsics and 6DoF poses. To this end, we propose NeRF$--$, with three contributions: First, we show that the camera parameters can be jointly optimised as learnable parameters with NeRF training, through a photometric reconstruction; Second, to benchmark the camera parameter estimation and the quality of novel view renderings, we introduce a new dataset of path-traced synthetic scenes, termed as Blender Forward-Facing Dataset (BLEFF); Third, we conduct extensive analyses to understand the training behaviours under various camera motions, and show that in most scenarios, the joint optimisation pipeline can recover accurate camera parameters and achieve comparable novel view synthesis quality as those trained with COLMAP pre-computed camera parameters. Our code and data are available at https://nerfmm.active.vision.

Theo W. Costain, Victor Adrian Prisacariu

Neural implicit representations have shown substantial improvements in efficiently storing 3D data, when compared to conventional formats. However, the focus of existing work has mainly been on storage and subsequent reconstruction. In this work, we show that training neural representations for reconstruction tasks alongside conventional tasks can produce more general encodings that admit equal quality reconstructions to single task training, whilst improving results on conventional tasks when compared to single task encodings. We reformulate the semantic segmentation task, creating a more representative task for implicit representation contexts, and through multi-task experiments on reconstruction, classification, and segmentation, show our approach learns feature rich encodings that admit equal performance for each task.

Marcelo Gennari do Nascimento, Theo W. Costain, Victor Adrian Prisacariu

We propose a novel method for neural network quantization that casts the neural architecture search problem as one of hyperparameter search to find non-uniform bit distributions throughout the layers of a CNN. We perform the search assuming a Multi-Task Gaussian Processes prior, which splits the problem to multiple tasks, each corresponding to different number of training epochs, and explore the space by sampling those configurations that yield maximum information. We then show that with significantly lower precision in the last layers we achieve a minimal loss of accuracy with appreciable memory savings. We test our findings on the CIFAR10 and ImageNet datasets using the VGG, ResNet and GoogLeNet architectures.

Xinghui Li, Kai Han, Shuda Li et al.

We tackle the problem of establishing dense pixel-wise correspondences between a pair of images. In this work, we introduce Dual-Resolution Correspondence Networks (DualRC-Net), to obtain pixel-wise correspondences in a coarse-to-fine manner. DualRC-Net extracts both coarse- and fine- resolution feature maps. The coarse maps are used to produce a full but coarse 4D correlation tensor, which is then refined by a learnable neighbourhood consensus module. The fine-resolution feature maps are used to obtain the final dense correspondences guided by the refined coarse 4D correlation tensor. The selected coarse-resolution matching scores allow the fine-resolution features to focus only on a limited number of possible matches with high confidence. In this way, DualRC-Net dramatically increases matching reliability and localisation accuracy, while avoiding to apply the expensive 4D convolution kernels on fine-resolution feature maps. We comprehensively evaluate our method on large-scale public benchmarks including HPatches, InLoc, and Aachen Day-Night. It achieves the state-of-the-art results on all of them.

Zirui Wang, Shuda Li, Henry Howard-Jenkins et al.

We present FlowNet3D++, a deep scene flow estimation network. Inspired by classical methods, FlowNet3D++ incorporates geometric constraints in the form of point-to-plane distance and angular alignment between individual vectors in the flow field, into FlowNet3D. We demonstrate that the addition of these geometric loss terms improves the previous state-of-art FlowNet3D accuracy from 57.85% to 63.43%. To further demonstrate the effectiveness of our geometric constraints, we propose a benchmark for flow estimation on the task of dynamic 3D reconstruction, thus providing a more holistic and practical measure of performance than the breakdown of individual metrics previously used to evaluate scene flow. This is made possible through the contribution of a novel pipeline to integrate point-based scene flow predictions into a global dense volume. FlowNet3D++ achieves up to a 15.0% reduction in reconstruction error over FlowNet3D, and up to a 35.2% improvement over KillingFusion alone. We will release our scene flow estimation code later.

Marcelo Gennari, Roger Fawcett, Victor Adrian Prisacariu

Quantization is a popular way of increasing the speed and lowering the memory usage of Convolution Neural Networks (CNNs). When labelled training data is available, network weights and activations have successfully been quantized down to 1-bit. The same cannot be said about the scenario when labelled training data is not available, e.g. when quantizing a pre-trained model, where current approaches show, at best, no loss of accuracy at 8-bit quantizations. We introduce DSConv, a flexible quantized convolution operator that replaces single-precision operations with their far less expensive integer counterparts, while maintaining the probability distributions over both the kernel weights and the outputs. We test our model as a plug-and-play replacement for standard convolution on most popular neural network architectures, ResNet, DenseNet, GoogLeNet, AlexNet and VGG-Net and demonstrate state-of-the-art results, with less than 1% loss of accuracy, without retraining, using only 4-bit quantization. We also show how a distillation-based adaptation stage with unlabelled data can improve results even further.

Victor Adrian Prisacariu, Olaf Kähler, Stuart Golodetz et al.

Volumetric models have become a popular representation for 3D scenes in recent years. One breakthrough leading to their popularity was KinectFusion, which focuses on 3D reconstruction using RGB-D sensors. However, monocular SLAM has since also been tackled with very similar approaches. Representing the reconstruction volumetrically as a TSDF leads to most of the simplicity and efficiency that can be achieved with GPU implementations of these systems. However, this representation is memory-intensive and limits applicability to small-scale reconstructions. Several avenues have been explored to overcome this. With the aim of summarizing them and providing for a fast, flexible 3D reconstruction pipeline, we propose a new, unifying framework called InfiniTAM. The idea is that steps like camera tracking, scene representation and integration of new data can easily be replaced and adapted to the user's needs. This report describes the technical implementation details of InfiniTAM v3, the third version of our InfiniTAM system. We have added various new features, as well as making numerous enhancements to the low-level code that significantly improve our camera tracking performance. The new features that we expect to be of most interest are (i) a robust camera tracking module; (ii) an implementation of Glocker et al.'s keyframe-based random ferns camera relocaliser; (iii) a novel approach to globally-consistent TSDF-based reconstruction, based on dividing the scene into rigid submaps and optimising the relative poses between them; and (iv) an implementation of Keller et al.'s surfel-based reconstruction approach.

Victor Adrian Prisacariu, Olaf Kähler, Ming Ming Cheng et al.

Volumetric models have become a popular representation for 3D scenes in recent years. One of the breakthroughs leading to their popularity was KinectFusion, where the focus is on 3D reconstruction using RGB-D sensors. However, monocular SLAM has since also been tackled with very similar approaches. Representing the reconstruction volumetrically as a truncated signed distance function leads to most of the simplicity and efficiency that can be achieved with GPU implementations of these systems. However, this representation is also memory-intensive and limits the applicability to small scale reconstructions. Several avenues have been explored for overcoming this limitation. With the aim of summarizing them and providing for a fast and flexible 3D reconstruction pipeline, we propose a new, unifying framework called InfiniTAM. The core idea is that individual steps like camera tracking, scene representation and integration of new data can easily be replaced and adapted to the needs of the user. Along with the framework we also provide a set of components for scalable reconstruction: two implementations of camera trackers, based on RGB data and on depth data, two representations of the 3D volumetric data, a dense volume and one based on hashes of subblocks, and an optional module for swapping subblocks in and out of the typically limited GPU memory.