This paper presents a novel paradigm for building scalable 3D generative models utilizing pre-trained video diffusion models. The primary obstacle in developing foundation 3D generative models is the limited availability of 3D data. Unlike images, texts, or videos, 3D data are not readily accessible and are difficult to acquire. This results in a significant disparity in scale compared to the vast quantities of other types of data. To address this issue, we propose using a video diffusion model, trained with extensive volumes of text, images, and videos, as a knowledge source for 3D data. By unlocking its multi-view generative capabilities through fine-tuning, we generate a large-scale synthetic multi-view dataset to train a feed-forward 3D generative model. The proposed model, VFusion3D, trained on nearly 3M synthetic multi-view data, can generate a 3D asset from a single image in seconds and achieves superior performance when compared to current SOTA feed-forward 3D generative models, with users preferring our results over 70% of the time.
Most text-to-3D generators build upon off-the-shelf text-to-image models trained on billions of images. They use variants of Score Distillation Sampling (SDS), which is slow, somewhat unstable, and prone to artifacts. A mitigation is to fine-tune the 2D generator to be multi-view aware, which can help distillation or can be combined with reconstruction networks to output 3D objects directly. In this paper, we further explore the design space of text-to-3D models. We significantly improve multi-view generation by considering video instead of image generators. Combined with a 3D reconstruction algorithm which, by using Gaussian splatting, can optimize a robust image-based loss, we directly produce high-quality 3D outputs from the generated views. Our new method, IM-3D, reduces the number of evaluations of the 2D generator network 10-100x, resulting in a much more efficient pipeline, better quality, fewer geometric inconsistencies, and higher yield of usable 3D assets.
We introduce Replay, a collection of multi-view, multi-modal videos of humans interacting socially. Each scene is filmed in high production quality, from different viewpoints with several static cameras, as well as wearable action cameras, and recorded with a large array of microphones at different positions in the room. Overall, the dataset contains over 4000 minutes of footage and over 7 million timestamped high-resolution frames annotated with camera poses and partially with foreground masks. The Replay dataset has many potential applications, such as novel-view synthesis, 3D reconstruction, novel-view acoustic synthesis, human body and face analysis, and training generative models. We provide a benchmark for training and evaluating novel-view synthesis, with two scenarios of different difficulty. Finally, we evaluate several baseline state-of-the-art methods on the new benchmark.
We present a method for fast 3D reconstruction and real-time rendering of dynamic humans from monocular videos with accompanying parametric body fits. Our method can reconstruct a dynamic human in less than 3h using a single GPU, compared to recent state-of-the-art alternatives that take up to 72h. These speedups are obtained by using a lightweight deformation model solely based on linear blend skinning, and an efficient factorized volumetric representation for modeling the shape and color of the person in canonical pose. Moreover, we propose a novel local ray marching rendering which, by exploiting standard GPU hardware and without any baking or conversion of the radiance field, allows visualizing the neural human on a mobile VR device at 40 frames per second with minimal loss of visual quality. Our experimental evaluation shows superior or competitive results with state-of-the art methods while obtaining large training speedup, using a simple model, and achieving real-time rendering.
We present MAV3D (Make-A-Video3D), a method for generating three-dimensional dynamic scenes from text descriptions. Our approach uses a 4D dynamic Neural Radiance Field (NeRF), which is optimized for scene appearance, density, and motion consistency by querying a Text-to-Video (T2V) diffusion-based model. The dynamic video output generated from the provided text can be viewed from any camera location and angle, and can be composited into any 3D environment. MAV3D does not require any 3D or 4D data and the T2V model is trained only on Text-Image pairs and unlabeled videos. We demonstrate the effectiveness of our approach using comprehensive quantitative and qualitative experiments and show an improvement over previously established internal baselines. To the best of our knowledge, our method is the first to generate 3D dynamic scenes given a text description.
Spatial self-attention layers, in the form of Non-Local blocks, introduce long-range dependencies in Convolutional Neural Networks by computing pairwise similarities among all possible positions. Such pairwise functions underpin the effectiveness of non-local layers, but also determine a complexity that scales quadratically with respect to the input size both in space and time. This is a severely limiting factor that practically hinders the applicability of non-local blocks to even moderately sized inputs. Previous works focused on reducing the complexity by modifying the underlying matrix operations, however in this work we aim to retain full expressiveness of non-local layers while keeping complexity linear. We overcome the efficiency limitation of non-local blocks by framing them as special cases of 3rd order polynomial functions. This fact enables us to formulate novel fast Non-Local blocks, capable of reducing the complexity from quadratic to linear with no loss in performance, by replacing any direct computation of pairwise similarities with element-wise multiplications. The proposed method, which we dub as "Poly-NL", is competitive with state-of-the-art performance across image recognition, instance segmentation, and face detection tasks, while having considerably less computational overhead.
We present To The Point (TTP), a method for reconstructing 3D objects from a single image using 2D to 3D correspondences learned from weak supervision. We recover a 3D shape from a 2D image by first regressing the 2D positions corresponding to the 3D template vertices and then jointly estimating a rigid camera transform and non-rigid template deformation that optimally explain the 2D positions through the 3D shape projection. By relying on 3D-2D correspondences we use a simple per-sample optimization problem to replace CNN-based regression of camera pose and non-rigid deformation and thereby obtain substantially more accurate 3D reconstructions. We treat this optimization as a differentiable layer and train the whole system in an end-to-end manner. We report systematic quantitative improvements on multiple categories and provide qualitative results comprising diverse shape, pose and texture prediction examples. Project website: https://fkokkinos.github.io/to_the_point/.
Monocular 3D reconstruction of articulated object categories is challenging due to the lack of training data and the inherent ill-posedness of the problem. In this work we use video self-supervision, forcing the consistency of consecutive 3D reconstructions by a motion-based cycle loss. This largely improves both optimization-based and learning-based 3D mesh reconstruction. We further introduce an interpretable model of 3D template deformations that controls a 3D surface through the displacement of a small number of local, learnable handles. We formulate this operation as a structured layer relying on mesh-laplacian regularization and show that it can be trained in an end-to-end manner. We finally introduce a per-sample numerical optimisation approach that jointly optimises over mesh displacements and cameras within a video, boosting accuracy both for training and also as test time post-processing. While relying exclusively on a small set of videos collected per category for supervision, we obtain state-of-the-art reconstructions with diverse shapes, viewpoints and textures for multiple articulated object categories.
Microscopy is a powerful visualization tool in biology, enabling the study of cells, tissues, and the fundamental biological processes. Yet, the observed images of the objects at the micro-scale suffer from two major inherent distortions: the blur caused by the diffraction of light, and the background noise caused by the imperfections of the imaging detectors. The latter is especially severe in fluorescence and in confocal microscopes, which are known for operating at the low photon count with the Poisson noise statistics. Restoration of such images is usually accomplished by image deconvolution, with the nature of the noise statistics taken into account, and by solving an optimization problem given some prior information about the underlying data (i.e., regularization). In this work, we propose a unifying framework of algorithms for Poisson image deblurring and denoising. The algorithms are based on deep learning techniques for the design of learnable regularizers paired with an appropriate optimization scheme. Our extensive experimentation line showcases that the proposed approach achieves superior quality of image reconstruction and beats the solutions that rely on deep learning or on the optimization schemes alone. Moreover, several implementations of the proposed framework demonstrate competitive performance at a low computational complexity, which is of high importance for real-time imaging applications.