ECCV 2024 Papers

"We present , a principled method for distributing Neural Radiance Fields (NeRFs) across multiple GPUs, thus enabling the training and rendering of NeRFs with an arbitrarily large capacity. We begin by revisiting existing multi-GPU approaches, which decompose large scenes into multiple independently trained NeRFs [?, ?, ?], and identify several fundamental issues with these methods that hinder improvements in reconstruction quality as additional computational resources (GPUs) are used in training. remedies these issues and enables the training and rendering of NeRFs with an arbitrary number of parameters by simply using more hardware. At the core of our method lies a novel distributed training and rendering formulation, which is mathematically equivalent to the classic single-GPU case and minimizes communication between GPUs. By unlocking NeRFs with arbitrarily large parameter counts, our approach is the first to reveal multi-GPU scaling laws for NeRFs, showing improvements in reconstruction quality with larger parameter counts and speed improvements with more GPUs. We demonstrate the effectiveness of on a wide variety of datasets, including the largest open-source dataset to date, MatrixCity [?], containing 258K images covering a 25km2 city area. Visit our webpage at https: //research.nvidia.com/labs/toronto-ai/nerfxl/ for code and videos."

"Neural fields excel in computer vision and robotics due to their ability to understand the 3D visual world such as inferring semantics, geometry, and dynamics. Given the capabilities of neural fields in densely representing a 3D scene from 2D images, we ask the question: Can we scale their self-supervised pretraining, specifically using masked autoencoders, to generate effective 3D representations from posed RGB images. Owing to the astounding success of extending transformers to novel data modalities, we employ standard 3D Vision Transformers to suit the unique formulation of NeRFs. We leverage NeRF’s volumetric grid as a dense input to the transformer, contrasting it with other 3D representations such as pointclouds where the information density can be uneven, and the representation is irregular. Due to the difficulty of applying masked autoencoders to an implicit representation, such as NeRF, we opt for extracting an explicit representation that canonicalizes scenes across domains by employing the camera trajectory for sampling. Our goal is made possible by masking random patches from NeRF’s radiance and density grid and employing a standard 3D Swin Transformer to reconstruct the masked patches. In doing so, the model can learn the semantic and spatial structure of complete scenes. We pretrain this representation at scale on our proposed curated posed-RGB data, totaling over 1.8 million images. Once pretrained, the encoder is used for effective 3D transfer learning. Our novel self-supervised pretraining for NeRFs, NeRF-MAE, scales remarkably well and improves performance on various challenging 3D tasks. Utilizing unlabeled posed 2D data for pretraining, NeRF-MAE significantly outperforms self-supervised 3D pretraining and NeRF scene understanding baselines on Front3D and ScanNet datasets with an absolute performance improvement of over 20% AP50 and 8% AP25 for 3D object detection. Project Page: magentanerf-mae.github.io"

"This work proposes NePhi, a generalizable neural deformation model which results in approximately diffeomorphic transformations. In contrast to the predominant voxel-based transformation fields used in learning-based registration approaches, NePhi represents deformations functionally, leading to great flexibility within the design space of memory consumption during training and inference, inference time, registration accuracy, as well as transformation regularity. Specifically, NePhi 1) requires less memory compared to voxel-based learning approaches, 2) improves inference speed by predicting latent codes, compared to current existing neural deformation based registration approaches that only rely on optimization, 3) improves accuracy via instance optimization, and 4) shows excellent deformation regularity which is highly desirable for medical image registration. We demonstrate the performance of NePhi on a 2D synthetic dataset as well as for real 3D medical image datasets (e.g., lungs and brains). Our results show that NePhi can match the accuracy of voxel-based representations in a single-resolution registration setting. For multi-resolution registration, our method matches the accuracy of current SOTA learning-based registration approaches with instance optimization while reducing memory requirements by a factor of five. Our code is available at https://github.com/uncbiag/NePhi."

"Navigation instruction generation, which requires embodied agents to describe the navigation routes, has been of great interest in robotics and human-computer interaction. Existing studies directly map the sequence of 2D perspective observations to route descriptions. Though straightforward, they overlook the geometric information and object semantics of the 3D environment. To address these challenges, we propose BEVInstructor, which incorporates Bird’s Eye View (BEV) features into Multi-Modal Large Language Models (MLLMs) for instruction generation. Specifically, BEVInstructor constructs a Perspective-BEV Visual Encoder for the comprehension of 3D environments through fusing BEV and perspective features. To leverage the powerful language capabilities of MLLMs, the fused representations are used as visual prom-pts for MLLMs, and perspective-BEV prompt tuning is proposed for parameter-efficient updating. Based on the perspective-BEV prompts, BEVInstructor further adopts an instance-guided iterative refinement pipeline, which improves the instructions in a progressive manner. BEVInstructor achieves impressive performance across diverse datasets (, R2R, REVERIE, and UrbanWalk)."

"This research investigates biases in text-to-image (TTI) models for the Indic languages widely spoken across India. It evaluates and compares the generative performance and cultural relevance of leading TTI models in these languages against their performance in English. Using the proposed IndicTTI benchmark, we comprehensively assess the performance of 30 Indic languages with two open-source diffusion models and two commercial generation APIs. The primary objective of this benchmark is to evaluate the support for Indic languages in these models and identify areas needing improvement. Given the linguistic diversity of 30 languages spoken by over 1.4 billion people, this benchmark aims to provide a detailed and insightful analysis of TTI models’ effectiveness within the Indic linguistic landscape. The data and code for the IndicTTI benchmark can be accessed at https://iab-rubric.org/resources/other-databases/ indictti."

"Capitalizing on the remarkable advancements in Large Language Models (LLMs), there is a burgeoning initiative to harness LLMs for instruction following robotic navigation. Such a trend underscores the potential of LLMs to generalize navigational reasoning and diverse language understanding. However, a significant discrepancy in agent performance is observed when integrating LLMs in the Vision-and-Language navigation (VLN) tasks compared to previous downstream specialist models. Furthermore, the inherent capacity of language to interpret and facilitate communication in agent interactions is often underutilized in these integrations. In this work, we strive to bridge the divide between VLN-specialized models and LLM-based navigation paradigms, while maintaining the interpretative prowess of LLMs in generating linguistic navigational reasoning. By aligning visual content in a frozen LLM, we encompass visual observation comprehension for LLMs and exploit a way to incorporate LLMs and navigation policy networks for effective action predictions and navigational reasoning. We demonstrate the data efficiency of the proposed methods and eliminate the gap between LM-based agents and state-of-the-art VLN specialists. The source code is available at https://github.com/GengzeZhou/NavGPT-2."

"Recent advancements in Novel View Synthesis (NVS) from a single image have produced impressive results by leveraging the generation capabilities of pre-trained Text-to-Image (T2I) models. However, previous NVS approaches require extra optimization to use other plug-and-play image generation modules such as ControlNet and LoRA, as they fine-tune the T2I parameters. In this study, we propose an efficient plug-and-play adaptation module, NVS-Adapter, that is compatible with existing plug-and-play modules without extensive fine-tuning. We introduce target views and reference view alignment to improve the geometric consistency of multi-view predictions. Experimental results demonstrate the compatibility of our NVS-Adapter with existing plug-and-play modules. Moreover, our NVS-Adapter shows superior performance over state-of-the-art methods on NVS benchmarks although it does not fine-tune billions of parameters of the pre-trained T2I models. The code and data are publicly available at postech-cvlab.github.io/nvsadapter/ 1 1 This work was done during the summer internship program at KakaoBrain."

"Discriminative models for object classification typically learn image-based representations that do not capture the compositional and 3D nature of objects. In this work, we show that explicitly integrating 3D compositional object representations into deep networks for image classification leads to a largely enhanced generalization in out-of-distribution scenarios. In particular, we introduce a novel architecture, referred to as , that consists of a feature extractor and a neural object volume for every target object class. Each neural object volume is a composition of 3D Gaussians that emit feature vectors. This compositional object representation allows for a highly robust and fast estimation of the object class by independently matching the features of the 3D Gaussians of each category to features extracted from an input image. Additionally, the object pose can be estimated via inverse rendering of the corresponding neural object volume. To enable the classification of objects, the neural features at each 3D Gaussian are trained discriminatively to be distinct from (i) the features of 3D Gaussians in other categories, (ii) features of other 3D Gaussians of the same object, and (iii) the background features. Our experiments show that offers intriguing advantages over standard architectures due to the 3D compositional structure of the object representation, namely: (1) An exceptional robustness across a spectrum of real-world and synthetic out-of-distribution shifts and (2) an enhanced human interpretability compared to standard models, all while maintaining real-time inference and a competitive accuracy on in-distribution data. Code and model can be found at /GenIntel/NOVUM."

"Modeling the physical contacts between the hand and object is standard for refining inaccurate hand poses and generating novel human grasp in 3D hand-object reconstruction. However, existing methods rely on geometric constraints that cannot be specified or controlled. This paper introduces a novel task of controllable 3D hand-object contact modeling with natural language descriptions. Challenges include i) the complexity of cross-modal modeling from language to contact, and ii) a lack of descriptive text for contact patterns. To address these issues, we propose NL2Contact, a model that generates controllable contacts by leveraging staged diffusion models. Given a language description of the hand and contact, NL2Contact generates realistic and faithful 3D hand-object contacts. To train the model, we build ContactDescribe, the first dataset with hand-centered contact descriptions. It contains multi-level and diverse descriptions generated by large language models based on carefully designed prompts (e.g., grasp action, grasp type, contact location, free finger status). We show applications of our model to grasp pose optimization and novel human grasp generation, both based on a textual contact description."

"Aligning a template to 3D human point clouds is a long-standing problem crucial for tasks like animation, reconstruction, and enabling supervised learning pipelines. Recent data-driven methods leverage predicted surface correspondences. However, they are not robust to varied poses, identities, or noise. In contrast, industrial solutions often rely on expensive manual annotations or multi-view capturing systems. Recently, neural fields have shown promising results. Still, their purely data-driven and extrinsic nature does not incorporate any guidance toward the target surface, often resulting in a trivial misalignment of the template registration. Currently, no method can be considered the standard for 3D Human registration, limiting the scalability of downstream applications. In this work, we propose a neural scalable registration method, , a pipeline that, for the first time, generalizes and scales across thousands of shapes and more than ten different data sources. Our essential contribution is , an ICP-style self-supervised task tailored to neural fields. takes a few seconds, is self-supervised, and works out of the box on pre-trained neural fields. combines with a localized neural field trained on a large MoCap dataset, achieving the state of the art over public benchmarks. The release of our code and checkpoints provides a powerful tool useful for many downstream tasks like dataset alignments, cleaning, or asset animation."

"This paper presents NGP-RT, a novel approach for enhancing the rendering speed of Instant-NGP to achieve real-time novel view synthesis. As a classic NeRF-based method, Instant-NGP stores implicit features in multi-level grids or hash tables and applies a shallow MLP to convert the implicit features into explicit colors and densities. Although it achieves fast training speed, there is still a lot of room for improvement in its rendering speed due to the per-point MLP executions for implicit multi-level feature aggregation, especially for real-time applications. To address this challenge, our proposed NGP-RT explicitly stores colors and densities as hash features, and leverages a lightweight attention mechanism to disambiguate the hash collisions instead of using computationally intensive MLP. At the rendering stage, NGP-RT incorporates a pre-computed occupancy distance grid into the ray marching strategy to inform the distance to the nearest occupied voxel, thereby reducing the number of marching points and global memory access. Experimental results show that on the challenging Mip-NeRF 360 dataset, NGP-RT achieves better rendering quality than previous NeRF-based methods, achieving 108 fps at 1080p resolution on a single Nvidia RTX 3090 GPU. Our approach is promising for NeRF-based real-time applications that require efficient and high-quality rendering."

"Recently, Handwritten Mathematical Expression Recognition (HMER) has gained considerable attention in pattern recognition for its diverse applications in document understanding. Current methods typically approach HMER as an image-to-sequence generation task within an autoregressive (AR) encoder-decoder framework. However, these approaches suffer from several drawbacks: 1) a lack of overall language context, limiting information utilization beyond the current decoding step; 2) error accumulation during AR decoding; and 3) slow decoding speed. To tackle these problems, this paper makes the first attempt to build a novel bottom-up Non-AutoRegressive Modeling approach for HMER, called NAMER. NAMER comprises a Visual Aware Tokenizer (VAT) and a Parallel Graph Decoder (PGD). Initially, the VAT tokenizes visible symbols and local relations at a coarse level. Subsequently, the PGD refines all tokens and establishes connectivities in parallel, leveraging comprehensive visual and linguistic contexts. Experiments on CROHME 2014/2016/2019 and HME100K datasets demonstrate that NAMER not only outperforms the current state-of-the-art (SOTA) methods on ExpRate by 1.93%/2.35%/1.49%/0.62%, but also achieves significant speedups of 13.7× and 6.7× faster in decoding time and overall FPS, proving the effectiveness and efficiency of NAMER."

"Understanding complex scenes at multiple levels of abstraction remains a formidable challenge in computer vision. To address this, we introduce Nested Neural Feature Fields (), a novel approach that employs hierarchical supervision to learn a single feature field, wherein different dimensions within the same high-dimensional feature encode scene properties at varying granularities. Our method allows for a flexible definition of hierarchies, tailored to either the physical dimensions or semantics or both, thereby enabling a comprehensive and nuanced understanding of scenes. We leverage a 2D class-agnostic segmentation model to provide semantically meaningful pixel groupings at arbitrary scales in the image space, and query the CLIP vision-encoder to obtain language-aligned embeddings for each of these segments. Our proposed hierarchical supervision method then assigns different nested dimensions of the feature field to distill the CLIP embeddings using deferred volumetric rendering at varying physical scales, creating a coarse-to-fine representation. Extensive experiments show that our approach outperforms the state-of-the-art feature field distillation methods on tasks such as open-vocabulary 3D segmentation and localization, demonstrating the effectiveness of the learned nested feature field."

"Perspective distortion (PD) causes unprecedented changes in shape, size, orientation, angles, and other spatial relationships of visual concepts in images. Precisely estimating camera intrinsic and extrinsic parameters is a challenging task that prevents synthesizing perspective distortion. Non-availability of dedicated training data poses a critical barrier to developing robust computer vision methods. Additionally, distortion correction methods make other computer vision tasks a multi-step approach and lack performance. In this work, we propose mitigating perspective distortion (MPD) by employing a fine-grained parameter control on a specific family of Möbius transform to model real-world distortion without estimating camera intrinsic and extrinsic parameters and without the need for actual distorted data. Also, we present a dedicated perspectively distorted benchmark dataset, ImageNet-PD, to benchmark the robustness of deep learning models against this new dataset. The proposed method outperforms existing benchmarks, ImageNet-E and ImageNet-X. Additionally, it significantly improves performance on ImageNet-PD while consistently performing on standard data distribution. Notably, our method shows improved performance on three PD-affected real-world applications—crowd counting, fisheye image recognition, and person re-identification—and one PD-affected challenging CV task: object detection. The source code, dataset, and models are available on the project webpage at https://prakashchhipa.github.io/projects/mpd."

"Recent large-scale vision-language models (VLMs) have demonstrated remarkable capabilities in understanding and generating textual descriptions for visual content. However, these models lack an understanding of user-specific concepts. In this work, we take a first step toward the personalization of VLMs, enabling them to learn and reason over user-provided concepts. For example, we explore whether these models can learn to recognize you in an image and communicate what you are doing, tailoring the model to reflect your personal experiences and relationships. To effectively recognize a variety of user-specific concepts, we augment the VLM with external concept heads that function as toggles for the model, enabling the VLM to identify the presence of specific target concepts in a given image. Having recognized the concept, we learn a new concept embedding in the intermediate feature space of the VLM. This embedding is tasked with guiding the language model to naturally integrate the target concept in its generated response. We apply our technique to BLIP-2 and LLaVA for personalized image captioning and further show its applicability for personalized visual question-answering. Our experiments demonstrate our ability to generalize to unseen images of learned concepts while preserving the model behavior on unrelated inputs. Code and data will be made available upon acceptance."

"Visual acoustic matching (VAM) is pivotal for enhancing the immersive experience, and the task of dereverberation is effective in improving audio intelligibility. Existing methods treat each task independently, overlooking the inherent reciprocity between them. Moreover, these methods depend on paired training data, which is challenging to acquire, impeding the utilization of extensive unpaired data. In this paper, we introduce MVSD, a mutual learning framework based on diffusion models. MVSD considers the two tasks symmetrically, exploiting the reciprocal relationship to facilitate learning from inverse tasks and overcome data scarcity. Furthermore, we employ the diffusion model as foundational conditional converters to circumvent the training instability and over-smoothing drawbacks of conventional GAN architectures. Specifically, MVSD employs two converters: one for VAM called reverberator and one for dereverberation called dereverberator. The dereverberator judges whether the reverberation audio generated by reverberator sounds like being in the conditional visual scenario, and vice versa. By forming a closed loop, these two converters can generate informative feedback signals to optimize the inverse tasks, even with easily acquired one-way unpaired data. Extensive experiments on two standard benchmarks, , SoundSpaces-Speech and Acoustic AVSpeech, exhibit that our framework can improve the performance of the reverberator and dereverberator and better match specified visual scenarios."

"Detection pre-training methods for the DETR series detector have been extensively studied in natural scenes, e.g., DETReg. However, the detection pre-training remains unexplored in remote sensing scenes. In existing pre-training methods, alignment between object embeddings extracted from a pre-trained backbone and detector features is significant. However, due to differences in feature extraction methods, a pronounced feature discrepancy still exists and hinders the pre-training performance. The remote sensing images with complex environments and more densely distributed objects exacerbate the discrepancy. In this work, we propose a novel Mutually optimizing pre-training framework for remote sensing object Detection, dubbed as MutDet. In MutDet, we propose a systemic solution against this challenge. Firstly, we propose a mutual enhancement module, which fuses the object embeddings and detector features bidirectionally in the last encoder layer, enhancing their information interaction. Secondly, contrastive alignment loss is employed to guide this alignment process softly and simultaneously enhances detector features’ discriminativity. Finally, we design an auxiliary siamese head to mitigate the task gap arising from the introduction of enhancement module. Comprehensive experiments on various settings show new state-of-the-art transfer performance. The improvement is particularly pronounced when data quantity is limited. When using 10 % of the DIOR-R data, MutDet improves DetReg by 6.1% in AP50 . Codes and models are available at: https://github.com/floatingstarZ/MutDet."

"Developing self-supervised learning (SSL) models that can learn universal and transferable representations of H&E gigapixel whole-slide images (WSIs) is becoming increasingly valuable in computational pathology. These models hold the potential to advance critical tasks such as few-shot classification, slide retrieval, and patient stratification. Existing approaches for slide representation learning extend the principles of SSL from small images (e.g., 224×224 patches) to entire slides, usually by aligning two different augmentations (or views) of the slide. Yet the resulting representation remains constrained by the limited clinical and biological diversity of the views. Instead, we postulate that slides stained with multiple markers, such as immunohistochemistry, can be used as different views to form a rich task-agnostic training signal. To this end, we introduce , a multimodal pretraining strategy for slide representation learning. is trained with a dual global-local cross-stain alignment objective on large cohorts of breast cancer samples (N=4,211 WSIs across five stains) and kidney transplant samples (N=12,070 WSIs across four stains). We demonstrate the quality of slide representations learned by on various downstream evaluations, ranging from morphological and molecular classification to prognostic prediction, comprising 21 tasks using 7,299 WSIs from multiple medical centers. Code at https://github.com/mahmoodlab/MADELEIN"

"Contemporary color difference (CD) measures for photographic images typically operate by comparing co-located pixels, patches in a “perceptually uniform” color space, or features in a learned latent space. Consequently, these measures inadequately capture the human color perception of misaligned image pairs, which are prevalent in digital photography (, the same scene captured by different smartphones). In this paper, we describe a perceptual CD measure based on the multiscale sliced Wasserstein distance, which facilitates efficient comparisons between non-local patches of similar color and structure. This aligns with the modern understanding of color perception, where color and structure are inextricably interdependent as a unitary process of perceptual organization. Meanwhile, our method is easy to implement and training-free. Experimental results indicate that our CD measure performs favorably in assessing CDs in photographic images, and consistently surpasses competing models in the presence of image misalignment. Additionally, we empirically verify that our measure functions as a metric in the mathematical sense, and show its promise as a loss function for image and video color transfer tasks. The code is available at https://github.com/real-hjq/MS-SWD."