We present Mesh2NeRF, an approach to derive ground-truth radiance fields from textured meshes for 3D generation tasks. Many 3D generative approaches represent 3D scenes as radiance fields for training. Their ground-truth radiance fields are usually fitted from multi-view renderings from a large-scale synthetic 3D dataset, which often results in artifacts due to occlusions or under-fitting issues. In Mesh2NeRF, we propose an analytic solution to directly obtain ground-truth radiance fields from 3D meshes, characterizing the density field with an occupancy function featuring a defined surface thickness, and determining view-dependent color through a reflection function considering both the mesh and environment lighting. Mesh2NeRF extracts accurate radiance fields which provides direct supervision for training generative NeRFs and single scene representation. We validate the effectiveness of Mesh2NeRF across various tasks, achieving a noteworthy 3.12dB improvement in PSNR for view synthesis in single scene representation on the ABO dataset, a 0.69 PSNR enhancement in the single-view conditional generation of ShapeNet Cars, and notably improved mesh extraction from NeRF in the unconditional generation of Objaverse Mugs.
Image matching is a fundamental computer vision problem. While learning-based methods achieve state-of-the-art performance on existing benchmarks, they generalize poorly to in-the-wild images. Such methods typically need to train separate models for different scene types and are impractical when the scene type is unknown in advance. One of the underlying problems is the limited scalability of existing data construction pipelines, which limits the diversity of standard image matching datasets. To address this problem, we propose GIM, a self-training framework for learning a single generalizable model based on any image matching architecture using internet videos, an abundant and diverse data source. Given an architecture, GIM first trains it on standard domain-specific datasets and then combines it with complementary matching methods to create dense labels on nearby frames of novel videos. These labels are filtered by robust fitting, and then enhanced by propagating them to distant frames. The final model is trained on propagated data with strong augmentations. We also propose ZEB, the first zero-shot evaluation benchmark for image matching. By mixing data from diverse domains, ZEB can thoroughly assess the cross-domain generalization performance of different methods. Applying GIM consistently improves the zero-shot performance of 3 state-of-the-art image matching architectures; with 50 hours of YouTube videos, the relative zero-shot performance improves by 8.4%-18.1%. GIM also enables generalization to extreme cross-domain data such as Bird Eye View (BEV) images of projected 3D point clouds (Fig. 1(c)). More importantly, our single zero-shot model consistently outperforms domain-specific baselines when evaluated on downstream tasks inherent to their respective domains. The video presentation is available at https://www.youtube.com/watch?v=FU_MJLD8LeY.
This work presents the network architecture EVP (Enhanced Visual Perception). EVP builds on the previous work VPD which paved the way to use the Stable Diffusion network for computer vision tasks. We propose two major enhancements. First, we develop the Inverse Multi-Attentive Feature Refinement (IMAFR) module which enhances feature learning capabilities by aggregating spatial information from higher pyramid levels. Second, we propose a novel image-text alignment module for improved feature extraction of the Stable Diffusion backbone. The resulting architecture is suitable for a wide variety of tasks and we demonstrate its performance in the context of single-image depth estimation with a specialized decoder using classification-based bins and referring segmentation with an off-the-shelf decoder. Comprehensive experiments conducted on established datasets show that EVP achieves state-of-the-art results in single-image depth estimation for indoor (NYU Depth v2, 11.8% RMSE improvement over VPD) and outdoor (KITTI) environments, as well as referring segmentation (RefCOCO, 2.53 IoU improvement over ReLA). The code and pre-trained models are publicly available at https://github.com/Lavreniuk/EVP.
Online continual learning, the process of training models on streaming data, has gained increasing attention in recent years. However, a critical aspect often overlooked is the label delay, where new data may not be labeled due to slow and costly annotation processes. We introduce a new continual learning framework with explicit modeling of the label delay between data and label streams over time steps. In each step, the framework reveals both unlabeled data from the current time step $t$ and labels delayed with $d$ steps, from the time step $t-d$. In our extensive experiments amounting to 1060 GPU days, we show that merely augmenting the computational resources is insufficient to tackle this challenge. Our findings underline a notable performance decline when solely relying on labeled data when the label delay becomes significant. More surprisingly, when using state-of-the-art SSL and TTA techniques to utilize the newer, unlabeled data, they fail to surpass the performance of a na\"ive method that simply trains on the delayed supervised stream. To this end, we introduce a simple, efficient baseline that rehearses from the labeled memory samples that are most similar to the new unlabeled samples. This method bridges the accuracy gap caused by label delay without significantly increasing computational complexity. We show experimentally that our method is the least affected by the label delay factor and in some cases successfully recovers the accuracy of the non-delayed counterpart. We conduct various ablations and sensitivity experiments, demonstrating the effectiveness of our approach.
Event cameras have emerged as a promising vision sensor in recent years due to their unparalleled temporal resolution and dynamic range. While registration of 2D RGB images to 3D point clouds is a long-standing problem in computer vision, no prior work studies 2D-3D registration for event cameras. To this end, we propose E2PNet, the first learning-based method for event-to-point cloud registration. The core of E2PNet is a novel feature representation network called Event-Points-to-Tensor (EP2T), which encodes event data into a 2D grid-shaped feature tensor. This grid-shaped feature enables matured RGB-based frameworks to be easily used for event-to-point cloud registration, without changing hyper-parameters and the training procedure. EP2T treats the event input as spatio-temporal point clouds. Unlike standard 3D learning architectures that treat all dimensions of point clouds equally, the novel sampling and information aggregation modules in EP2T are designed to handle the inhomogeneity of the spatial and temporal dimensions. Experiments on the MVSEC and VECtor datasets demonstrate the superiority of E2PNet over hand-crafted and other learning-based methods. Compared to RGB-based registration, E2PNet is more robust to extreme illumination or fast motion due to the use of event data. Beyond 2D-3D registration, we also show the potential of EP2T for other vision tasks such as flow estimation, event-to-image reconstruction and object recognition. The source code can be found at: https://github.com/Xmu-qcj/E2PNet.
We release MiDaS v3.1 for monocular depth estimation, offering a variety of new models based on different encoder backbones. This release is motivated by the success of transformers in computer vision, with a large variety of pretrained vision transformers now available. We explore how using the most promising vision transformers as image encoders impacts depth estimation quality and runtime of the MiDaS architecture. Our investigation also includes recent convolutional approaches that achieve comparable quality to vision transformers in image classification tasks. While the previous release MiDaS v3.0 solely leverages the vanilla vision transformer ViT, MiDaS v3.1 offers additional models based on BEiT, Swin, SwinV2, Next-ViT and LeViT. These models offer different performance-runtime tradeoffs. The best model improves the depth estimation quality by 28% while efficient models enable downstream tasks requiring high frame rates. We also describe the general process for integrating new backbones. A video summarizing the work can be found at https://youtu.be/UjaeNNFf9sE and the code is available at https://github.com/isl-org/MiDaS.
This paper proposes a novel online evaluation protocol for Test Time Adaptation (TTA) methods, which penalizes slower methods by providing them with fewer samples for adaptation. TTA methods leverage unlabeled data at test time to adapt to distribution shifts. Though many effective methods have been proposed, their impressive performance usually comes at the cost of significantly increased computation budgets. Current evaluation protocols overlook the effect of this extra computation cost, affecting their real-world applicability. To address this issue, we propose a more realistic evaluation protocol for TTA methods, where data is received in an online fashion from a constant-speed data stream, thereby accounting for the method's adaptation speed. We apply our proposed protocol to benchmark several TTA methods on multiple datasets and scenarios. Extensive experiments shows that, when accounting for inference speed, simple and fast approaches can outperform more sophisticated but slower methods. For example, SHOT from 2020 outperforms the state-of-the-art method SAR from 2023 under our online setting. Our online evaluation protocol emphasizes the need for developing TTA methods that are efficient and applicable in realistic settings.
We present a visual-inertial depth estimation pipeline that integrates monocular depth estimation and visual-inertial odometry to produce dense depth estimates with metric scale. Our approach performs global scale and shift alignment against sparse metric depth, followed by learning-based dense alignment. We evaluate on the TartanAir and VOID datasets, observing up to 30% reduction in inverse RMSE with dense scale alignment relative to performing just global alignment alone. Our approach is especially competitive at low density; with just 150 sparse metric depth points, our dense-to-dense depth alignment method achieves over 50% lower iRMSE over sparse-to-dense depth completion by KBNet, currently the state of the art on VOID. We demonstrate successful zero-shot transfer from synthetic TartanAir to real-world VOID data and perform generalization tests on NYUv2 and VCU-RVI. Our approach is modular and is compatible with a variety of monocular depth estimation models. Video: https://youtu.be/IMwiKwSpshQ Code: https://github.com/isl-org/VI-Depth
This paper tackles the problem of depth estimation from a single image. Existing work either focuses on generalization performance disregarding metric scale, i.e. relative depth estimation, or state-of-the-art results on specific datasets, i.e. metric depth estimation. We propose the first approach that combines both worlds, leading to a model with excellent generalization performance while maintaining metric scale. Our flagship model, ZoeD-M12-NK, is pre-trained on 12 datasets using relative depth and fine-tuned on two datasets using metric depth. We use a lightweight head with a novel bin adjustment design called metric bins module for each domain. During inference, each input image is automatically routed to the appropriate head using a latent classifier. Our framework admits multiple configurations depending on the datasets used for relative depth pre-training and metric fine-tuning. Without pre-training, we can already significantly improve the state of the art (SOTA) on the NYU Depth v2 indoor dataset. Pre-training on twelve datasets and fine-tuning on the NYU Depth v2 indoor dataset, we can further improve SOTA for a total of 21% in terms of relative absolute error (REL). Finally, ZoeD-M12-NK is the first model that can jointly train on multiple datasets (NYU Depth v2 and KITTI) without a significant drop in performance and achieve unprecedented zero-shot generalization performance to eight unseen datasets from both indoor and outdoor domains. The code and pre-trained models are publicly available at https://github.com/isl-org/ZoeDepth .
Humans naturally exploit haptic feedback during contact-rich tasks like loading a dishwasher or stocking a bookshelf. Current robotic systems focus on avoiding unexpected contact, often relying on strategically placed environment sensors. Recently, contact-exploiting manipulation policies have been trained in simulation and deployed on real robots. However, they require some form of real-world adaptation to bridge the sim-to-real gap, which might not be feasible in all scenarios. In this paper we train a contact-exploiting manipulation policy in simulation for the contact-rich household task of loading plates into a slotted holder, which transfers without any fine-tuning to the real robot. We investigate various factors necessary for this zero-shot transfer, like time delay modeling, memory representation, and domain randomization. Our policy transfers with minimal sim-to-real gap and significantly outperforms heuristic and learnt baselines. It also generalizes to plates of different sizes and weights. Demonstration videos and code are available at https://sites.google.com/view/compliant-object-insertion.