Neural surface representation has demonstrated remarkable success in the areas of novel view synthesis and 3D reconstruction. However, assessing the geometric quality of 3D reconstructions in the absence of ground truth mesh remains a significant challenge, due to its rendering-based optimization process and entangled learning of appearance and geometry with photometric losses. In this paper, we present a novel framework, i.e, GURecon, which establishes a geometric uncertainty field for the neural surface based on geometric consistency. Different from existing methods that rely on rendering-based measurement, GURecon models a continuous 3D uncertainty field for the reconstructed surface, and is learned by an online distillation approach without introducing real geometric information for supervision. Moreover, in order to mitigate the interference of illumination on geometric consistency, a decoupled field is learned and exploited to finetune the uncertainty field. Experiments on various datasets demonstrate the superiority of GURecon in modeling 3D geometric uncertainty, as well as its plug-and-play extension to various neural surface representations and improvement on downstream tasks such as incremental reconstruction.

Predicting click-through rates is crucial in various fields, including online advertising and recommendation systems. The key to improving the performance of CTR prediction lies in learning a robust user representation, particularly by analyzing their historical behaviors. Previous studies usually model behavior sequences through attention-based sequence models or graph-based methods, which usually struggle to explore diverse latent interests or accurately model user behaviors. Moreover, this challenge is exacerbated when users' historical behaviors are sparse, a common issue in real-world business-to-business (B2B) e-commerce scenarios. In this paper, we propose a novel Graph-Enhanced Interest Network (GEIN) to capture users' latent intents and facilitate the sequential learning of sparse behavior sequences. Specifically, we first construct a hierarchical item-intent heterogeneous graph to enrich the representation of sparse behaviors using diverse information from graphs. Next, we build a user-level behavior interest factor graph to accurately capture user interests. Additionally, a contrastive learning mechanism is incorporated to mitigate the negative robustness impacts caused by sparsity. Extensive experiments on real-world datasets demonstrate that our proposed GEIN outperforms a wide range of state-of-the-art methods. Furthermore, online A/B testing also confirms the superiority of GEIN over competing baselines in a real-world production environment.

User prediction aims to identify the user who show potential value in e-commerce platforms. Existing user prediction algorithms primarily focus on distinguishing between potential user, those likely to purchase a specific product in the future, and non-potential user, those unlikely to do so. However, these algorithms often overlook a valuable group named conditional potential user in non-potential group. Conditional potential user, initially categorized as non-potential, can convert into potential user under specific converted condition (e.g., product trial, coupon receipt, or certification attainment). Additionally, conventional user prediction models do not offer insights into the associated converted condition and provide classification results only. Furthermore, the lack of explainability in user prediction methods limits their practical utility. To tackle these challenges, this paper proposes a definitive concept for conditional potential users and introduces an innovative user prediction framework, the Conditional Potential User Mining (CPUM). Addressing the challenge of predicting conditional potential users, CPUM employs a Gold Standard Classifier to ascertain users’ intentions, thereby identifying their interest in target products under converted conditions and enabling the prediction of conditional potential users. Moreover, CPUM adopts an explainable surrogate model with an attention mechanism to boost explainability. This model is designed to mirror the decision-making capacity of the Gold Standard Classifier and provides importance explanations for different user interactions. This enables precise identification of converted conditions in conditional potential users, thus enhancing explainability. The effectiveness of this method is substantiated by experimental results derived from extensive e-commerce datasets from Alibaba and Amazon, demonstrating its cutting-edge performance and applicability. • Introduces the concept of conditional potential users in e-commerce. • Launches CPUM framework for accurate prediction of conditional potential users. • Enhances model explainability with a surrogate model and attention mechanism. • Validates CPUM’s effectiveness with Amazon and Alibaba datasets. • Offers a scalable user prediction model adaptable to various e-commerce platforms.

Due to the ability to synthesize high-quality novel views, Neural Radiance Fields (NeRF) has been recently exploited to improve visual localization in a known environment. However, the existing methods mostly utilize NeRF for data augmentation to improve the regression model training, and their performances on novel viewpoints and appearances are still limited due to the lack of geometric constraints. In this paper, we propose a novel visual localization framework, i.e., PNeRFLoc, based on a unified point-based representation. On one hand, PNeRFLoc supports the initial pose estimation by matching 2D and 3D feature points as traditional structure-based methods; on the other hand, it also enables pose refinement with novel view synthesis using rendering-based optimization. Specifically, we propose a novel feature adaption module to close the gaps between the features for visual localization and neural rendering. To improve the efficacy and efficiency of neural rendering-based optimization, we also developed an efficient rendering-based framework with a warping loss function. Extensive experiments demonstrate that PNeRFLoc performs the best on the synthetic dataset when the 3D NeRF model can be well learned, and significantly outperforms all the NeRF-boosted localization methods with on-par SOTA performance on the real-world benchmark localization datasets. Project webpage: https://zju3dv.github.io/PNeRFLoc/.

To cater to users' desire for an immersive browsing experience, numerous e-commerce platforms provide various recommendation scenarios, with a focus on Trigger-Induced Recommendation (TIR) tasks. However, the majority of current TIR methods heavily rely on the trigger item to understand user intent, lacking a higher-level exploration and exploitation of user intent (e.g., popular items and complementary items), which may result in an overly convergent understanding of users' short-term intent and can be detrimental to users' long-term purchasing experiences. Moreover, users' short-term intent shows uncertainty and is affected by various factors such as browsing context and historical behaviors, which poses challenges to user intent modeling. To address these challenges, we propose a novel model called Deep Uncertainty Intent Network (DUIN), comprising three essential modules: i) Explicit Intent Exploit Module extracting explicit user intent using the contrastive learning paradigm; ii) Latent Intent Explore Module exploring latent user intent by leveraging the multi-view relationships between items; iii) Intent Uncertainty Measurement Module offering a distributional estimation and capturing the uncertainty associated with user intent. Experiments on three real-world datasets demonstrate the superior performance of DUIN compared to existing baselines. Notably, DUIN has been deployed across all TIR scenarios in our e-commerce platform, with online A/B testing results conclusively validating its superiority.

To solve the cold-start issue that cold items have no historical interactions to obtain collaborative feature as their representation, existing methods often represent them totally based on content feature obtained from inherent content (i.e., image, video and attributes). However, these methods will lose efficacy in representing the cold item whose inherent content is much different from that of warm items, when the cold item first appears in the test stage. In this paper, we propose a mutual information assisted graph convolution network (MIGCN), which represents the cold item by simultaneously considering its inherent content and its related users’ feature captured by pair-wise mutual information (PMI). In addition, for improving the overall performance, a joint objective function that contains a relation loss and a similarity error loss is employed to achieve a better trade-off of the representation similarity between warm and cold items. Experiments conducted on the Amazon dataset demonstrate the superiority of our method, as compared with the state-of-the-art methods.

Click-through rate (CTR) prediction is one of the fundamental tasks in the industry, especially in e-commerce, social media, and streaming media. It directly impacts website revenues, user satisfaction, and user retention. However, real-world production platforms often encompass various domains to cater for diverse customer needs. Traditional CTR prediction models struggle in multi-domain recommendation scenarios, facing challenges of data sparsity and disparate data distributions across domains. Existing multi-domain recommendation approaches introduce specific-domain modules for each domain, which partially address these issues but often significantly increase model parameters and lead to insufficient training. In this paper, we propose a Multi-domain Low-Rank Adaptive network (MLoRA) for CTR prediction, where we introduce a specialized LoRA module for each domain. This approach enhances the model’s performance in multi-domain CTR prediction tasks and is able to be applied to various deep-learning models. We evaluate the proposed method on several multi-domain datasets. Experimental results demonstrate our MLoRA approach achieves a significant improvement compared with state-of-the-art baselines. Furthermore, we deploy it in the production environment of the Alibaba.COM 1. The online A/B testing results indicate the superiority and flexibility in real-world production environments. The code of our MLoRA is publicly available 2.

The recommendation has been playing a key role in many industries, e.g., e-commerce, streaming media, social media, etc. Recently, a new recommendation scenario, called Trigger-Induced Recommendation (TIR), where users are able to explicitly express their instant interests via trigger items, is emerging as an essential role in many e-commerce platforms, e.g., Alibaba.com and Amazon. Without explicitly modeling the user's instant interest, traditional recommendation methods usually obtain sub-optimal results in TIR. Even though there are a few methods considering the trigger and target items simultaneously to solve this problem, they still haven't taken into account temporal information of user behaviors, the dynamic change of user instant interest when the user scrolls down and the interactions between the trigger and target items. To tackle these problems, we propose a novel method -- Deep Evolutional Instant Interest Network (DEI2N), for click-through rate prediction in TIR scenarios. Specifically, we design a User Instant Interest Modeling Layer to predict the dynamic change of the intensity of instant interest when the user scrolls down. Temporal information is utilized in user behavior modeling. Moreover, an Interaction Layer is introduced to learn better interactions between the trigger and target items. We evaluate our method on several offline and real-world industrial datasets. Experimental results show that our proposed DEI2N outperforms state-of-the-art baselines. In addition, online A/B testing demonstrates the superiority over the existing baseline in real-world production environments.

Standard visual localization methods build a priori 3D model of a scene which is used to establish correspondences against the 2D keypoints in a query image. Storing these pre-built 3D scene models can be prohibitively expensive for large-scale environments, especially on mobile devices with limited storage and communication bandwidth. We design a novel framework that compresses a scene while still maintaining localization accuracy. The scene is compressed in three stages: first, the database frames are clustered using pairwise co-visibility information. Then, a learned point selection module prunes the points in each cluster taking into account the final pose estimation accuracy. In the final stage, the features of the selected points are further compressed using learned quantization. Query image registration is done using only the compressed scene points. To the best of our knowledge, we are the first to propose learned scene compression for visual localization. We also demonstrate the effectiveness and efficiency of our method on various outdoor datasets where it can perform accurate localization with low memory consumption.

This paper presents an end-to-end neural network for multiple rotation averaging in SfM. Due to the manifold constraint of rotations, conventional methods usually take two separate steps involving spanning tree based initialization and iterative nonlinear optimization respectively. These methods can suffer from bad initializations due to the noisy spanning tree or outliers in input relative rotations. To handle these problems, we propose to integrate initialization and optimization together in an unified graph neural network via a novel differentiable multi-source propagation module. Specifically, our network utilizes the image context and geometric cues in feature correspondences to reduce the impact of outliers. Furthermore, unlike the methods that utilize the spanning tree to initialize orientations according to a single reference node in a top-down manner, our net-work initializes orientations according to multiple sources while utilizing information from all neighbors in a differentiable way. More importantly, our end-to-end formulation also enables iterative re-weighting of input relative orientations at test time to improve the accuracy of the final estimation by minimizing the impact of outliers. We demonstrate the effectiveness of our method on two real-world datasets, achieving state-of-the-art performance.

A bipartite network is a graph structure where nodes are from two distinct domains and only inter-domain interactions exist as edges. A large number of network embedding methods exist to learn vectorial node representations from general graphs with both homogeneous and heterogeneous node and edge types, including some that can specifically model the distinct properties of bipartite networks. However, these methods are inadequate to model multiplex bipartite networks (e.g., in e-commerce), that have multiple types of interactions (e.g., click, inquiry, and buy) and node attributes. Most real-world multiplex bipartite networks are also sparse and have imbalanced node distributions that are challenging to model. In this paper, we develop an unsupervised Dual HyperGraph Convolutional Network (DualHGCN) model that scalably transforms the multiplex bipartite network into two sets of homogeneous hypergraphs and uses spectral hypergraph convolutional operators, along with intra- and inter-message passing strategies to promote information exchange within and across domains, to learn effective node embeddings. We benchmark DualHGCN using four real-world datasets on link prediction and node classification tasks. Our extensive experiments demonstrate that DualHGCN significantly outperforms state-of-the-art methods, and is robust to varying sparsity levels and imbalanced node distributions.

Click-through rate prediction plays an important role in many fields, such as recommender and advertising systems. It is one of the crucial parts to improve user experience and increase industry revenue. Recently, several deep learning-based models are successfully applied to this area. Some existing studies further model user representation based on user historical behavior sequence, in order to capture dynamic and evolving interests. We observe that users usually have multiple interests at a time and the latent dominant interest is expressed by the behavior. The switch of latent dominant interest results in the behavior changes. Thus, modeling and tracking latent multiple interests would be beneficial. In this paper, we propose a novel method named as Deep Multi-Interest Network (DMIN) which models user's latent multiple interests for click-through rate prediction task. Specifically, we design a Behavior Refiner Layer using multi-head self-attention to capture better user historical item representations. Then the Multi-Interest Extractor Layer is applied to extract multiple user interests. We evaluate our method on three real-world datasets. Experimental results show that the proposed DMIN outperforms various state-of-the-art baselines in terms of click-through rate prediction task.

This paper presents a scene agnostic neural architecture for camera localization, where model parameters and scenes are independent from each other.Despite recent advancement in learning based methods, most approaches require training for each scene one by one, not applicable for online applications such as SLAM and robotic navigation, where a model must be built on-the-fly.Our approach learns to build a hierarchical scene representation and predicts a dense scene coordinate map of a query RGB image on-the-fly given an arbitrary scene. The 6D camera pose of the query image can be estimated with the predicted scene coordinate map. Additionally, the dense prediction can be used for other online robotic and AR applications such as obstacle avoidance. We demonstrate the effectiveness and efficiency of our method on both indoor and outdoor benchmarks, achieving state-of-the-art performance.

This paper presents a novel polarimetric dense monocular SLAM (PDMS) algorithm based on a polarization camera. The algorithm exploits both photometric and polarimetric light information to produce more accurate and complete geometry. The polarimetric information allows us to recover the azimuth angle of surface normals from each video frame to facilitate dense reconstruction, especially at textureless or specular regions. There are two challenges in our approach: 1) surface azimuth angles from the polarization camera are very noisy; and 2) we need a near real-time solution for SLAM. Previous successful methods on polarimetric multi-view stereo are offline and require manually pre-segmented object masks to suppress the effects of erroneous angle information along boundaries. Our fully automatic approach efficiently iterates azimuth-based depth propagations, two-view depth consistency check, and depth optimization to produce a depthmap in real-time, where all the algorithmic steps are carefully designed to enable a GPU implementation. To our knowledge, this paper is the first to propose a photometric and polarimetric method for dense SLAM. We have qualitatively and quantitatively evaluated our algorithm against a few of competing methods, demonstrating the superior performance on various indoor and outdoor scenes.

Transcribing the singing voice into music notes is challenging due to pitch fluctuations such as portamenti and vibratos. This paper presents a probabilistic transcription method for monophonic sung melodies that explicitly accounts for these local pitch fluctuations. In the hierarchical Hidden Markov Model (HMM), an upper-level ergodic HMM handles the transitions between notes, and a lower-level left-to-right HMM handles the intra- and inter-note pitch fluctuations. The lower-level HMM employs the pitch dynamic model, which explicitly expresses the pitch curve characteristics as the observation likelihood over f0and Δf0using a compact parametric distribution. A histogram-based tuning frequency estimation method, and some post-processing heuristics to separate merged notes and to allocate spuriously detected short notes, improve the note recognition performance. With model parameters that support intuitions about singing behavior, the proposed method obtained encouraging results when evaluated on a published monophonic sung melody dataset, and compared with state-of-the-art methods.

Abstract In the paper, we present an interactive image‐segmentation method in the framework of graph cut, which incorporates not only traditional color and gradient constraints, but also a new type of texture constraint. Given an image with user‐input strokes, we first establish the color and texture prior models of the foreground/background. The texture prior model, which is key to establish the texture constraints, is represented by local binary patterns (LBP) histograms. Then, an energy function composed of color, gradient, and texture terms is formulated. At last, by using graph cut, we minimize the energy function to obtain the foreground. In the energy function, the color and gradient terms have similar forms with traditional methods. The texture term in the function is generated using a proposed randomized texton‐searching algorithm. First, the algorithm locates an approximately best representative texton for every unknown pixel as foreground and an approximately best one as background, through randomized searching. Second, it computes the LBP histograms of the two textons as the pixel's foreground/background texture descriptors, respectively. Finally, the distances between the descriptors and the foreground/background prior models are used to formulate the texture term. Experimental results demonstrate that our method outperforms traditional ones. Copyright © 2015 John Wiley & Sons, Ltd.

This letter presents an approach to extracting a target object interactively from a given pair of stereo images. First, a user marks a few parts of the object and background in either of the two views with strokes. The marked pixels are used to generate the prior models of the foreground and background. Second, a graph is constructed with constraints formulated by the priors of foreground/background, similarities between intraview neighbor pixels and correspondences between interview pixels. Third, two segments of the foreground are extracted from the two views by optimization of the graph via graph cut. Traditional methods generally define the priors and neighbor similarities in RGB space. Differently, the proposed method integrates disparity distributions of foreground/background to enrich the priors and defines the similarity metric between neighbor pixels in RGB-D space. The proposed method that utilizes RGB-D hybrid constraints generates stereo segments with accuracies higher than those obtained by state-of-the-art methods.