We propose LookinGood^{\pi}, a novel neural re-rendering approach that is aimed to (1) improve the rendering quality of the low-quality reconstructed results from human performance capture system in real-time; (2) improve the generalization ability of the neural rendering network on unseen people. Our key idea is to utilize the rendered image of reconstructed geometry as the guidance to assist the prediction of person-specific details from few reference images, thus enhancing the re-rendered result. In light of this, we design a two-branch network. A coarse branch is designed to fix some artifacts (i.e. holes, noise) and obtain a coarse version of the rendered input, while a detail branch is designed to predict "correct" details from the warped references. The guidance of the rendered image is realized by blending features from two branches effectively in the training of the detail branch, which improves both the warping accuracy and the details' fidelity. We demonstrate that our method outperforms state-of-the-art methods at producing high-fidelity images on unseen people.
相關內容
Convenient 4D modeling of human-object interactions is essential for numerous applications. However, monocular tracking and rendering of complex interaction scenarios remain challenging. In this paper, we propose Instant-NVR, a neural approach for instant volumetric human-object tracking and rendering using a single RGBD camera. It bridges traditional non-rigid tracking with recent instant radiance field techniques via a multi-thread tracking-rendering mechanism. In the tracking front-end, we adopt a robust human-object capture scheme to provide sufficient motion priors. We further introduce a separated instant neural representation with a novel hybrid deformation module for the interacting scene. We also provide an on-the-fly reconstruction scheme of the dynamic/static radiance fields via efficient motion-prior searching. Moreover, we introduce an online key frame selection scheme and a rendering-aware refinement strategy to significantly improve the appearance details for online novel-view synthesis. Extensive experiments demonstrate the effectiveness and efficiency of our approach for the instant generation of human-object radiance fields on the fly, notably achieving real-time photo-realistic novel view synthesis under complex human-object interactions.
Predicting athletes' performance has relied mostly on statistical data. Besides the traditional data, various types of data, including video, have become available. However, it is challenging to use them for deep learning, especially when the size of the athletes' dataset is small. This research proposes a feature-selection strategy based on the criteria used by insightful people, which could improve ML performance. Our ML model employs features selected by people who correctly evaluated the athletes' future performance. We tested out a strategy to predict the LPGA players' next day performance using their interview video. We asked study participants to predict the players' next day score after watching the interviews and asked why. Using combined features of the facial landmarks' movements, derived from the participants, and meta-data showed a better F1-score than using each feature separately. This study suggests that the human-in-the-loop model could improve algorithms' performance with small-dataset.
3D semantic segmentation is a critical task in many real-world applications, such as autonomous driving, robotics, and mixed reality. However, the task is extremely challenging due to ambiguities coming from the unstructured, sparse, and uncolored nature of the 3D point clouds. A possible solution is to combine the 3D information with others coming from sensors featuring a different modality, such as RGB cameras. Recent multi-modal 3D semantic segmentation networks exploit these modalities relying on two branches that process the 2D and 3D information independently, striving to maintain the strength of each modality. In this work, we first explain why this design choice is effective and then show how it can be improved to make the multi-modal semantic segmentation more robust to domain shift. Our surprisingly simple contribution achieves state-of-the-art performances on four popular multi-modal unsupervised domain adaptation benchmarks, as well as better results in a domain generalization scenario.
Multimodal reasoning, an area of artificial intelligence that aims at make inferences from multimodal signals such as vision, language and speech, has drawn more and more attention in recent years. People with different personalities may respond differently to the same situation. However, such individual personalities were ignored in the previous studies. In this work, we introduce a new Personality-aware Human-centric Multimodal Reasoning (Personality-aware HMR) task, and accordingly construct a new dataset based on The Big Bang Theory television shows, to predict the behavior of a specific person at a specific moment, given the multimodal information of its past and future moments. The Myers-Briggs Type Indicator (MBTI) was annotated and utilized in the task to represent individuals' personalities. We benchmark the task by proposing three baseline methods, two were adapted from the related tasks and one was newly proposed for our task. The experimental results demonstrate that personality can effectively improve the performance of human-centric multimodal reasoning. To further solve the lack of personality annotation in real-life scenes, we introduce an extended task called Personality-predicted HMR, and propose the corresponding methods, to predict the MBTI personality at first, and then use the predicted personality to help multimodal reasoning. The experimental results show that our method can accurately predict personality and achieves satisfactory multimodal reasoning performance without relying on personality annotations.
Fashion illustration is used by designers to communicate their vision and to bring the design idea from conceptualization to realization, showing how clothes interact with the human body. In this context, computer vision can thus be used to improve the fashion design process. Differently from previous works that mainly focused on the virtual try-on of garments, we propose the task of multimodal-conditioned fashion image editing, guiding the generation of human-centric fashion images by following multimodal prompts, such as text, human body poses, and garment sketches. We tackle this problem by proposing a new architecture based on latent diffusion models, an approach that has not been used before in the fashion domain. Given the lack of existing datasets suitable for the task, we also extend two existing fashion datasets, namely Dress Code and VITON-HD, with multimodal annotations collected in a semi-automatic manner. Experimental results on these new datasets demonstrate the effectiveness of our proposal, both in terms of realism and coherence with the given multimodal inputs. Source code and collected multimodal annotations will be publicly released at: //github.com/aimagelab/multimodal-garment-designer.
Recent 2D-to-3D human pose estimation works tend to utilize the graph structure formed by the topology of the human skeleton. However, we argue that this skeletal topology is too sparse to reflect the body structure and suffer from serious 2D-to-3D ambiguity problem. To overcome these weaknesses, we propose a novel graph convolution network architecture, Hierarchical Graph Networks (HGN). It is based on denser graph topology generated by our multi-scale graph structure building strategy, thus providing more delicate geometric information. The proposed architecture contains three sparse-to-fine representation subnetworks organized in parallel, in which multi-scale graph-structured features are processed and exchange information through a novel feature fusion strategy, leading to rich hierarchical representations. We also introduce a 3D coarse mesh constraint to further boost detail-related feature learning. Extensive experiments demonstrate that our HGN achieves the state-of-the art performance with reduced network parameters. Code is released at //github.com/qingshi9974/BMVC2021-Hierarchical-Graph-Networks-for-3D-Human-Pose-Estimation.
During the first part of life, the brain develops while it learns through a process called synaptogenesis. The neurons, growing and interacting with each other, create synapses. However, eventually the brain prunes those synapses. While previous work focused on learning and pruning independently, in this work we propose a biologically plausible model that, thanks to a combination of Hebbian learning and pruning, aims to simulate the synaptogenesis process. In this way, while learning how to solve the task, the agent translates its experience into a particular network structure. Namely, the network structure builds itself during the execution of the task. We call this approach Self-building Neural Network (SBNN). We compare our proposed SBNN with traditional neural networks (NNs) over three classical control tasks from OpenAI. The results show that our model performs generally better than traditional NNs. Moreover, we observe that the performance decay while increasing the pruning rate is smaller in our model than with NNs. Finally, we perform a validation test, testing the models over tasks unseen during the learning phase. In this case, the results show that SBNNs can adapt to new tasks better than the traditional NNs, especially when over $80\%$ of the weights are pruned.
Contrastive loss has been increasingly used in learning representations from multiple modalities. In the limit, the nature of the contrastive loss encourages modalities to exactly match each other in the latent space. Yet it remains an open question how the modality alignment affects the downstream task performance. In this paper, based on an information-theoretic argument, we first prove that exact modality alignment is sub-optimal in general for downstream prediction tasks. Hence we advocate that the key of better performance lies in meaningful latent modality structures instead of perfect modality alignment. To this end, we propose three general approaches to construct latent modality structures. Specifically, we design 1) a deep feature separation loss for intra-modality regularization; 2) a Brownian-bridge loss for inter-modality regularization; and 3) a geometric consistency loss for both intra- and inter-modality regularization. Extensive experiments are conducted on two popular multi-modal representation learning frameworks: the CLIP-based two-tower model and the ALBEF-based fusion model. We test our model on a variety of tasks including zero/few-shot image classification, image-text retrieval, visual question answering, visual reasoning, and visual entailment. Our method achieves consistent improvements over existing methods, demonstrating the effectiveness and generalizability of our proposed approach on latent modality structure regularization.
Human-centric perception plays a vital role in vision and graphics. But their data annotations are prohibitively expensive. Therefore, it is desirable to have a versatile pre-train model that serves as a foundation for data-efficient downstream tasks transfer. To this end, we propose the Human-Centric Multi-Modal Contrastive Learning framework HCMoCo that leverages the multi-modal nature of human data (e.g. RGB, depth, 2D keypoints) for effective representation learning. The objective comes with two main challenges: dense pre-train for multi-modality data, efficient usage of sparse human priors. To tackle the challenges, we design the novel Dense Intra-sample Contrastive Learning and Sparse Structure-aware Contrastive Learning targets by hierarchically learning a modal-invariant latent space featured with continuous and ordinal feature distribution and structure-aware semantic consistency. HCMoCo provides pre-train for different modalities by combining heterogeneous datasets, which allows efficient usage of existing task-specific human data. Extensive experiments on four downstream tasks of different modalities demonstrate the effectiveness of HCMoCo, especially under data-efficient settings (7.16% and 12% improvement on DensePose Estimation and Human Parsing). Moreover, we demonstrate the versatility of HCMoCo by exploring cross-modality supervision and missing-modality inference, validating its strong ability in cross-modal association and reasoning.
Retrieving object instances among cluttered scenes efficiently requires compact yet comprehensive regional image representations. Intuitively, object semantics can help build the index that focuses on the most relevant regions. However, due to the lack of bounding-box datasets for objects of interest among retrieval benchmarks, most recent work on regional representations has focused on either uniform or class-agnostic region selection. In this paper, we first fill the void by providing a new dataset of landmark bounding boxes, based on the Google Landmarks dataset, that includes $94k$ images with manually curated boxes from $15k$ unique landmarks. Then, we demonstrate how a trained landmark detector, using our new dataset, can be leveraged to index image regions and improve retrieval accuracy while being much more efficient than existing regional methods. In addition, we further introduce a novel regional aggregated selective match kernel (R-ASMK) to effectively combine information from detected regions into an improved holistic image representation. R-ASMK boosts image retrieval accuracy substantially at no additional memory cost, while even outperforming systems that index image regions independently. Our complete image retrieval system improves upon the previous state-of-the-art by significant margins on the Revisited Oxford and Paris datasets. Code and data will be released.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191