Current feed-forward 3D/4D reconstruction systems rely on dense geometry and pose supervision -- expensive to obtain at scale and particularly scarce for dynamic real-world scenes. We present Flow3r, a framework that augments visual geometry learning with dense 2D correspondences (`flow') as supervision, enabling scalable training from unlabeled monocular videos. Our key insight is that the flow prediction module should be factored: predicting flow between two images using geometry latents from one and pose latents from the other. This factorization directly guides the learning of both scene geometry and camera motion, and naturally extends to dynamic scenes. In controlled experiments, we show that factored flow prediction outperforms alternative designs and that performance scales consistently with unlabeled data. Integrating factored flow into existing visual geometry architectures and training with ${\sim}800$K unlabeled videos, Flow3r achieves state-of-the-art results across eight benchmarks spanning static and dynamic scenes, with its largest gains on in-the-wild dynamic videos where labeled data is most scarce.
Image captioning has attracted ever-increasing research attention in the multimedia community. To this end, most cutting-edge works rely on an encoder-decoder framework with attention mechanisms, which have achieved remarkable progress. However, such a framework does not consider scene concepts to attend visual information, which leads to sentence bias in caption generation and defects the performance correspondingly. We argue that such scene concepts capture higher-level visual semantics and serve as an important cue in describing images. In this paper, we propose a novel scene-based factored…
From just a glance, humans can make rich predictions about the future state of a wide range of physical systems. On the other hand, modern approaches from engineering, robotics, and graphics are often restricted to narrow domains and require direct measurements of the underlying states. We introduce the Visual Interaction Network, a general-purpose model for learning the dynamics of a physical system from raw visual observations. Our model consists of a perceptual front-end based on convolutional neural networks and a dynamics predictor based on interaction networks. Through joint training, th…
Visual fiducial systems are a key component of many robotics and AR/VR applications for 6-DOF monocular relative pose estimation and target identification. This paper presents LFTag, a visual fiducial system based on topological detection and relative position data encoding which optimizes data density within spatial frequency constraints. The marker is constructed to resolve rotational ambiguity, which combined with the robust geometric and topological false positive rejection, allows all marker bits to be used for data. When compared to existing state-of-the-art square binary markers (AprilT…
Georgia Institute of Technology（佐治亚理工学院）；Apple（苹果公司）
Existing research on making sense of deep neural networks often focuses on neuron-level interpretation, which may not adequately capture the bigger picture of how concepts are collectively encoded by multiple neurons. We present NeuroCartography, an interactive system that scalably summarizes and visualizes concepts learned by neural networks. It automatically discovers and groups neurons that detect the same concepts, and describes how such neuron groups interact to form higher-level concepts and the subsequent predictions. NeuroCartography introduces two scalable summarization techniques: (1…
The University of Hong Kong（香港大学）；Tencent AI Lab（腾讯人工智能实验室）
Pretraining Vision Transformers (ViTs) has achieved great success in visual recognition. A following scenario is to adapt a ViT to various image and video recognition tasks. The adaptation is challenging because of heavy computation and memory storage. Each model needs an independent and complete finetuning process to adapt to different tasks, which limits its transferability to different visual domains. To address this challenge, we propose an effective adaptation approach for Transformer, namely AdaptFormer, which can adapt the pre-trained ViTs into many different image and video tasks effic…
University of Chile（智利大学）；Snap Research（Snap 研究院）；Impresee Inc.（Impresee 公司）
Visual attributes play an essential role in real applications based on image retrieval. For instance, the extraction of attributes from images allows an eCommerce search engine to produce retrieval results with higher precision. The traditional manner to build an attribute extractor is by training a convnet-based classifier with a fixed number of classes. However, this approach does not scale for real applications where the number of attributes changes frequently. Therefore in this work, we propose an approach for extracting visual attributes from images, leveraging the learned capability of t…
Scalable Audio-Visual Masked Autoencoders for Efficient Affective Video Facial Analysis
School of Computer Science and Technology, Xi’an Jiaotong University（西安交通大学计算机科学与技术学院）；School of Cyber Science and Engineering, Zhengzhou University（郑州大学网络科学与工程学院）；School of Software, Northwestern Polytechnical University（西北工业大学软件学院）；Institute of Advanced Technology, University of Science and Technology of China（中国科学技术大学先进技术研究院）；Inspur Group（Inspur集团）；Research Center for Space Computing System, Zhejiang Lab（浙江实验室空间计算系统研究中心）
针对情感视频面部分析数据有限及跨模态建模难题，提出可扩展音视频 MAE 模型 AVF-MAE++，通过双重掩码、相关性学习模块和三阶段训练，在17个数据集上取得SOTA。
We propose a novel, efficient, modular and scalable framework for content based visual media retrieval systems by leveraging the power of Deep Learning which is flexible to work both for images and videos conjointly and we also introduce an efficient comparison and filtering metric for retrieval. We put forward our findings from critical performance tests comparing our method to the predominant conventional approach to demonstrate the feasibility and efficiency of the proposed solution with best practices, possible improvements that may further augment the ability of retrieval architectures.
The recently proposed Visual image Transformers (ViT) with pure attention have achieved promising performance on image recognition tasks, such as image classification. However, the routine of the current ViT model is to maintain a full-length patch sequence during inference, which is redundant and lacks hierarchical representation. To this end, we propose a Hierarchical Visual Transformer (HVT) which progressively pools visual tokens to shrink the sequence length and hence reduces the computational cost, analogous to the feature maps downsampling in Convolutional Neural Networks (CNNs). It bri…
University of Chinese Academy of Sciences（中国科学院大学）
In this paper, we explore the redundancy in convolutional neural network, which scales with the complexity of vision tasks. Considering that many front-end visual systems are interested in only a limited range of visual targets, the removing of task-specified network redundancy can promote a wide range of potential applications. We propose a task-specified knowledge distillation algorithm to derive a simplified model with pre-set computation cost and minimized accuracy loss, which suits the resource constraint front-end systems well. Experiments on the MNIST and CIFAR10 datasets demonstrate th…
Recently, the self-supervised pre-training paradigm has shown great potential in leveraging large-scale unlabeled data to improve downstream task performance. However, increasing the scale of unlabeled pre-training data in real-world scenarios requires prohibitive computational costs and faces the challenge of uncurated samples. To address these issues, we build a task-specific self-supervised pre-training framework from a data selection perspective based on a simple hypothesis that pre-training on the unlabeled samples with similar distribution to the target task can bring substantial perform…
School of Computer Science and Electronic Engineering, University of Essex（埃塞克斯大学计算机科学与电子工程学院）；School of Electrical Engineering and Computer Science, Queensland University of Technology（昆士兰科技大学电气工程与计算机科学学院）；National Centre for Nuclear Robotics Flexible Partnership Fund（国家核机器人灵活伙伴关系基金）
State-of-the-art visual place recognition performance is currently being achieved utilizing deep learning based approaches. Despite the recent efforts in designing lightweight convolutional neural network based models, these can still be too expensive for the most hardware restricted robot applications. Low-overhead VPR techniques would not only enable platforms equipped with low-end, cheap hardware but also reduce computation on more powerful systems, allowing these resources to be allocated for other navigation tasks. In this work, our goal is to provide an algorithm of extreme compactness a…
PixelPilot: Scalable Vision-Language-Action Models for End-to-End Autonomous Driving
MoE Key Lab of Artificial Intelligence, Institute of AI, Shanghai Jiao Tong University（教育部人工智能重点实验室，上海交通大学人工智能研究院）；Central Research Institute, Huawei（华为中央研究院）
This paper introduces a new fundamental characteristic, \ie, the dynamic range, from real-world metric tools to deep visual recognition. In metrology, the dynamic range is a basic quality of a metric tool, indicating its flexibility to accommodate various scales. Larger dynamic range offers higher flexibility. In visual recognition, the multiple scale problem also exist. Different visual concepts may have different semantic scales. For example, ``Animal'' and ``Plants'' have a large semantic scale while ``Elk'' has a much smaller one. Under a small semantic scale, two different elks may look q…
Vision-RWKV: Efficient and Scalable Visual Perception with RWKV-Like Architectures
The Chinese University of Hong Kong（香港中文大学）；Shanghai AI Laboratory（上海人工智能实验室）；Fudan University（复旦大学）；Nanjing University（南京大学）；Tsinghua University（清华大学）；SenseTime Research（商汤科技研究院）
Customizing Visual Emotion Evaluation for MLLMs: An Open-vocabulary, Multifaceted, and Scalable Approach
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；VCIP & TMCC & DISSec, College of Computer Science, Nankai University（南开大学计算机学院）；Department of Psychological and Cognitive Sciences, Tsinghua University（清华大学心理学与认知科学系）；University of Chinese Academy of Sciences（中国科学院大学）
Understanding visual reality involves acquiring common-sense knowledge about countless regularities in the visual world, e.g., how illumination alters the appearance of objects in a scene, and how motion changes their apparent spatial relationship. These regularities are hard to label for training supervised machine learning algorithms; consequently, algorithms need to learn these regularities from the real world in an unsupervised way. We present a novel network meta-architecture that can learn world dynamics from raw, continuous video. The components of this network can be implemented using …
A Unified and Scalable Membership Inference Method for Visual Self-supervised Encoder via Part-aware Capability
Key Lab of High Confidence Software Technologies (Peking University)（高可信软件技术重点实验室（北京大学））；Ministry of Education（教育部）；School of Computer Science（计算机学院）；Peking University（北京大学）；Tsinghua University（清华大学）
Visualization requirements in Forensic Lucid have to do with different levels of case knowledge abstraction, representation, aggregation, as well as the operational aspects as the final long-term goal of this proposal. It encompasses anything from the finer detailed representation of hierarchical contexts to Forensic Lucid programs, to the documented evidence and its management, its linkage to programs, to evaluation, and to the management of GIPSY software networks. This includes an ability to arbitrarily switch between those views combined with usable multimodal interaction. The purpose is t…
University of Maryland（马里兰大学）；MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）；Google Research（谷歌研究院）
We introduce CAN, a simple, efficient and scalable method for self-supervised learning of visual representations. Our framework is a minimal and conceptually clean synthesis of (C) contrastive learning, (A) masked autoencoders, and (N) the noise prediction approach used in diffusion models. The learning mechanisms are complementary to one another: contrastive learning shapes the embedding space across a batch of image samples; masked autoencoders focus on reconstruction of the low-frequency spatial correlations in a single image sample; and noise prediction encourages the reconstruction of the…
City University of Hong Kong（香港城市大学）；University of Chinese Academy of Sciences（中国科学院大学）；Peking University（北京大学）
It plays a fundamental role to compactly represent the visual information towards the optimization of the ultimate utility in myriad visual data centered applications. With numerous approaches proposed to efficiently compress the texture and visual features serving human visual perception and machine intelligence respectively, much less work has been dedicated to studying the interactions between them. Here we investigate the integration of feature and texture compression, and show that a universal and collaborative visual information representation can be achieved in a hierarchical way. In pa…
Bootstrap Dynamic-Aware 3D Visual Representation for Scalable Robot Learning
Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Shenzhen University（深圳大学）；Beijing Jiaotong University（北京交通大学）；Central South University（中南大学）
ScalSelect: Scalable Training-Free Multimodal Data Selection for Efficient Visual Instruction Tuning
East China Normal University（华东师范大学）；Zhongguancun Academy（中关村学院）；The Hong Kong Polytechnic University（香港理工大学）；Harbin Institute of Technology（哈尔滨工业大学）；Huazhong University of Science and Technology（华中科技大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Zhongguancun Institute of Artificial Intelligence（中关村人工智能研究院）
Crab$^{+}$: A Scalable and Unified Audio-Visual Scene Understanding Model with Explicit Cooperation
Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学耿丽人工智能学院）；Institute of Artificial Intelligence of China Telecom (TeleAI)（中国电信人工智能研究院）；AI Technology Center, Online Video Business Unit, Tencent PCG（腾讯PCG在线视频业务单元AI技术中心）；Hefei University of Technology（合肥工业大学）；The University of Hong Kong（香港大学）
UC Berkeley（加州大学伯克利分校）；ICSI（国际计算机科学研究所）；Microsoft Research Asia（微软亚洲研究院）
Current major approaches to visual recognition follow an end-to-end formulation that classifies an input image into one of the pre-determined set of semantic categories. Parametric softmax classifiers are a common choice for such a closed world with fixed categories, especially when big labeled data is available during training. However, this becomes problematic for open-set scenarios where new categories are encountered with very few examples for learning a generalizable parametric classifier. We adopt a non-parametric approach for visual recognition by optimizing feature embeddings instead o…
We study scalable and uniform understanding of facts in images. Existing visual recognition systems are typically modeled differently for each fact type such as objects, actions, and interactions. We propose a setting where all these facts can be modeled simultaneously with a capacity to understand unbounded number of facts in a structured way. The training data comes as structured facts in images, including (1) objects (e.g., $<$boy$>$), (2) attributes (e.g., $<$boy, tall$>$), (3) actions (e.g., $<$boy, playing$>$), and (4) interactions (e.g., $<$boy, riding, a horse $>$). Each fact has a sem…
In the last decade, video blogs (vlogs) have become an extremely popular method through which people express sentiment. The ubiquitousness of these videos has increased the importance of multimodal fusion models, which incorporate video and audio features with traditional text features for automatic sentiment detection. Multimodal fusion offers a unique opportunity to build models that learn from the full depth of expression available to human viewers. In the detection of sentiment in these videos, acoustic and video features provide clarity to otherwise ambiguous transcripts. In this paper, w…
