Jean-Marc Odobez

Chihiro Nakatani, Norimichi Ukita, Jean-Marc Odobez

This paper proposes an end-to-end shared attention estimation method via group detection. Most previous methods estimate shared attention (SA) without detecting the actual group of people focusing on it, or assume that there is a single SA point in a given image. These issues limit the applicability of SA detection in practice and impact performance. To address them, we propose to simultaneously achieve group detection and shared attention estimation using a two step process: (i) the generation of SA heatmaps relying on individual gaze attention heatmaps and group membership scalars estimated in a group inference; (ii) a refinement of the initial group memberships allowing to account for the initial SA heatmaps, and the final prediction of the SA heatmap. Experiments demonstrate that our method outperforms other methods in group detection and shared attention estimation. Additional analyses validate the effectiveness of the proposed components. Code: https://github.com/chihina/sagd-CVPRW2026.

Samy Tafasca, Anshul Gupta, Jean-Marc Odobez

Gaze behaviors such as eye-contact or shared attention are important markers for diagnosing developmental disorders in children. While previous studies have looked at some of these elements, the analysis is usually performed on private datasets and is restricted to lab settings. Furthermore, all publicly available gaze target prediction benchmarks mostly contain instances of adults, which makes models trained on them less applicable to scenarios with young children. In this paper, we propose the first study for predicting the gaze target of children and interacting adults. To this end, we introduce the ChildPlay dataset: a curated collection of short video clips featuring children playing and interacting with adults in uncontrolled environments (e.g. kindergarten, therapy centers, preschools etc.), which we annotate with rich gaze information. We further propose a new model for gaze target prediction that is geometrically grounded by explicitly identifying the scene parts in the 3D field of view (3DFoV) of the person, leveraging recent geometry preserving depth inference methods. Our model achieves state of the art results on benchmark datasets and ChildPlay. Furthermore, results show that looking at faces prediction performance on children is much worse than on adults, and can be significantly improved by fine-tuning models using child gaze annotations. Our dataset and models will be made publicly available.

Anshul Gupta, Samy Tafasca, Jean-Marc Odobez

Predicting where a person is looking is a complex task, requiring to understand not only the person's gaze and scene content, but also the 3D scene structure and the person's situation (are they manipulating? interacting or observing others? attentive?) to detect obstructions in the line of sight or apply attention priors that humans typically have when observing others. In this paper, we hypothesize that identifying and leveraging such priors can be better achieved through the exploitation of explicitly derived multimodal cues such as depth and pose. We thus propose a modular multimodal architecture allowing to combine these cues using an attention mechanism. The architecture can naturally be exploited in privacy-sensitive situations such as surveillance and health, where personally identifiable information cannot be released. We perform extensive experiments on the GazeFollow and VideoAttentionTarget public datasets, obtaining state-of-the-art performance and demonstrating very competitive results in the privacy setting case.

Samy Tafasca, Anshul Gupta, Jean-Marc Odobez

Gaze is a powerful form of non-verbal communication and social interaction that humans develop from an early age. As such, modeling this behavior is an important task that can benefit a broad set of application domains ranging from robotics to sociology. In particular, Gaze Following is defined as the prediction of the pixel-wise 2D location where a person in the image is looking. Prior efforts in this direction have focused primarily on CNN-based architectures to perform the task. In this paper, we introduce a novel transformer-based architecture for 2D gaze prediction. We experiment with 2 variants: the first one retains the same task formulation of predicting a gaze heatmap for one person at a time, while the second one casts the problem as a 2D point regression and allows us to perform multi-person gaze prediction with a single forward pass. This new architecture achieves state-of-the-art results on the GazeFollow and VideoAttentionTarget datasets. The code for this paper will be made publicly available.

Abid Ali, Rui Dai, Ashish Marisetty et al.

The computer vision community has explored dyadic interactions for atomic actions such as pushing, carrying-object, etc. However, with the advancement in deep learning models, there is a need to explore more complex dyadic situations such as loose interactions. These are interactions where two people perform certain atomic activities to complete a global action irrespective of temporal synchronisation and physical engagement, like cooking-together for example. Analysing these types of dyadic-interactions has several useful applications in the medical domain for social-skills development and mental health diagnosis. To achieve this, we propose a novel dual-path architecture to capture the loose interaction between two individuals. Our model learns global abstract features from each stream via a CNNs backbone and fuses them using a new Global-Layer-Attention module based on a cross-attention strategy. We evaluate our model on real-world autism diagnoses such as our Loose-Interaction dataset, and the publicly available Autism dataset for loose interactions. Our network achieves baseline results on the Loose-Interaction and SOTA results on the Autism datasets. Moreover, we study different social interactions by experimenting on a publicly available dataset i.e. NTU-RGB+D (interactive classes from both NTU-60 and NTU-120). We have found that different interactions require different network designs. We also compare a slightly different version of our method by incorporating time information to address tight interactions achieving SOTA results.

Arya Farkhondeh, Samy Tafasca, Jean-Marc Odobez

Hand-Object Interaction (HOI) is gaining significant attention, particularly with the creation of numerous egocentric datasets driven by AR/VR applications. However, third-person view HOI has received less attention, especially in terms of datasets. Most third-person view datasets are curated for action recognition tasks and feature pre-segmented clips of high-level daily activities, leaving a gap for in-the-wild datasets. To address this gap, we propose ChildPlay-Hand, a novel dataset that includes person and object bounding boxes, as well as manipulation actions. ChildPlay-Hand is unique in: (1) providing per-hand annotations; (2) featuring videos in uncontrolled settings with natural interactions, involving both adults and children; (3) including gaze labels from the ChildPlay-Gaze dataset for joint modeling of manipulations and gaze. The manipulation actions cover the main stages of an HOI cycle, such as grasping, holding or operating, and different types of releasing. To illustrate the interest of the dataset, we study two tasks: object in hand detection (OiH), i.e. if a person has an object in their hand, and manipulation stages (ManiS), which is more fine-grained and targets the main stages of manipulation. We benchmark various spatio-temporal and segmentation networks, exploring body vs. hand-region information and comparing pose and RGB modalities. Our findings suggest that ChildPlay-Hand is a challenging new benchmark for modeling HOI in the wild.

Abid Ali, Mahmoud Ali, Jean-Marc Odobez et al.

Autism Spectrum Disorder (ASD) is a diverse collection of neurobiological conditions marked by challenges in social communication and reciprocal interactions, as well as repetitive and stereotypical behaviors. Atypical behavior patterns in a long, untrimmed video can serve as biomarkers for children with ASD. In this paper, we propose a video-based weakly-supervised method that takes spatio-temporal features of long videos to learn typical and atypical behaviors for autism detection. On top of that, we propose a shallow TCN-MLP network, which is designed to further categorize the severity score. We evaluate our method on actual evaluation videos of children with autism collected and annotated (for severity score) by clinical professionals. Experimental results demonstrate the effectiveness of behavioral biomarkers that could help clinicians in autism spectrum analysis.

Hengfei Wang, Anshul Gupta, Pierre Vuillecard et al.

Vision-language models (VLMs) have rapidly evolved into general-purpose multimodal reasoners with strong zero-shot generalization. In this context, VLMs could greatly benefit the analysis of human gaze and attention, a central task in human behavior understanding that requires reasoning about the physical scene as well as the activity, interactions, and social context. However, the extent to which VLMs can reliably understand human gaze and related attentional behaviors remains largely unexplored. In this work, we present EyeVLM, a systematic evaluation framework for gaze understanding in VLMs across two complementary dimensions: tasks and models. To assess gaze understanding capabilities, we focus on two core tasks. The first, gaze following, i.e., predicting the 2D location where a person is looking, has a geometric and visual processing focus, requiring a precise understanding of the human face, attention direction, 3D scene structure, and spatial grounding of attended targets. The second, social gaze prediction, requires social and relational reasoning over multi-person interactions (e.g., mutual gaze and shared attention), and may benefit more from the LLM semantic reasoning capabilities within VLMs. Regarding models, EyeVLM evaluates these tasks in two ways: a zero-shot setting with a diverse set of state-of-the-art open- and closed-source VLMs, exploring different prompting strategies; and a fine-tuning approach based on task-specific QA pairs, studying the impact of model scale and data scale. As benchmarks, we rely on existing gaze understanding datasets and perform a systematic comparison with state-of-the-art purely visual models. Overall, our results show that current VLMs lack precise gaze understanding capabilities. While standard training helps reduce the gap with visual models, significant improvements are still needed.

Yihui Fu, Zhuoyuan Yao, Weipeng He et al.

The IEEE Spoken Language Technology Workshop (SLT) 2021 Alpha-mini Speech Challenge (ASC) is intended to improve research on keyword spotting (KWS) and sound source location (SSL) on humanoid robots. Many publications report significant improvements in deep learning based KWS and SSL on open source datasets in recent years. For deep learning model training, it is necessary to expand the data coverage to improve the robustness of model. Thus, simulating multi-channel noisy and reverberant data from single-channel speech, noise, echo and room impulsive response (RIR) is widely adopted. However, this approach may generate mismatch between simulated data and recorded data in real application scenarios, especially echo data. In this challenge, we open source a sizable speech, keyword, echo and noise corpus for promoting data-driven methods, particularly deep-learning approaches on KWS and SSL. We also choose Alpha-mini, a humanoid robot produced by UBTECH equipped with a built-in four-microphone array on its head, to record development and evaluation sets under the actual Alpha-mini robot application scenario, including noise as well as echo and mechanical noise generated by the robot itself for model evaluation. Furthermore, we illustrate the rules, evaluation methods and baselines for researchers to quickly assess their achievements and optimize their models.

Masoumeh Chapariniya, Jean-Marc Odobez, Volker Dellwo et al.

Avatar fingerprinting, i.e., verifying who drives a synthetic talking-head video rather than whether it is real, is a critical safeguard for authorized use of face-reenactment technology. Existing methods rely on a fixed, non-differentiable landmark extraction stage that prevents the fingerprinting model from being optimized end-to-end from raw pixels. We propose a preprocessing-free system built on a micro-expression-aware backbone operating on raw video frames, with inter-frame feature differencing as the core design principle: consecutive feature maps are subtracted in the learned deep feature space, so that temporally stable appearance dimensions contribute zero to the output while driver-specific motion dynamics are preserved. A controlled ablation on NVFAIR confirms that temporal motion accounts for the large majority of discriminative performance, and that raw appearance features actively degrade identity separation. Both the choice of backbone and the differencing principle are essential: differencing alone is insufficient when applied to a generic encoder, as appearance-dominated features collapse to near-identical representations across adjacent frames, while the micro-expression-aware F5C backbone retains measurable motion variation that the differencing operation can exploit. Without any external preprocessing, our model achieves an overall AUC of 0.877 on NVFAIR and matches or exceeds the landmark-based baseline on the majority of cross-generator pairs.

Anshul Gupta, Samy Tafasca, Arya Farkhondeh et al.

Gaze following and social gaze prediction are fundamental tasks providing insights into human communication behaviors, intent, and social interactions. Most previous approaches addressed these tasks separately, either by designing highly specialized social gaze models that do not generalize to other social gaze tasks or by considering social gaze inference as an ad-hoc post-processing of the gaze following task. Furthermore, the vast majority of gaze following approaches have proposed static models that can handle only one person at a time, therefore failing to take advantage of social interactions and temporal dynamics. In this paper, we address these limitations and introduce a novel framework to jointly predict the gaze target and social gaze label for all people in the scene. The framework comprises of: (i) a temporal, transformer-based architecture that, in addition to image tokens, handles person-specific tokens capturing the gaze information related to each individual; (ii) a new dataset, VSGaze, that unifies annotation types across multiple gaze following and social gaze datasets. We show that our model trained on VSGaze can address all tasks jointly, and achieves state-of-the-art results for multi-person gaze following and social gaze prediction.

Pierre Vuillecard, Jean-Marc Odobez

Accurate 3D gaze estimation in unconstrained real-world environments remains a significant challenge due to variations in appearance, head pose, occlusion, and the limited availability of in-the-wild 3D gaze datasets. To address these challenges, we introduce a novel Self-Training Weakly-Supervised Gaze Estimation framework (ST-WSGE). This two-stage learning framework leverages diverse 2D gaze datasets, such as gaze-following data, which offer rich variations in appearances, natural scenes, and gaze distributions, and proposes an approach to generate 3D pseudo-labels and enhance model generalization. Furthermore, traditional modality-specific models, designed separately for images or videos, limit the effective use of available training data. To overcome this, we propose the Gaze Transformer (GaT), a modality-agnostic architecture capable of simultaneously learning static and dynamic gaze information from both image and video datasets. By combining 3D video datasets with 2D gaze target labels from gaze following tasks, our approach achieves the following key contributions: (i) Significant state-of-the-art improvements in within-domain and cross-domain generalization on unconstrained benchmarks like Gaze360 and GFIE, with notable cross-modal gains in video gaze estimation; (ii) Superior cross-domain performance on datasets such as MPIIFaceGaze and Gaze360 compared to frontal face methods. Code and pre-trained models will be released to the community.

Masoumeh Chapariniya, Pierre Vuillecard, Jean-Marc Odobez et al.

This work investigates whether individuals can be identified solely through the pure dynamical components of their facial expressions, independent of static facial appearance. We leverage the FLAME 3D morphable model to achieve explicit disentanglement between facial shape and expression dynamics, extracting frame-by-frame parameters from conversational videos while retaining only expression and jaw coefficients. On the CANDOR dataset of 1,429 speakers in naturalistic conversations, our Conformer model with supervised contrastive learning achieves 61.14\%accuracy on 1,429-way classification -- 458 times above chance -- demonstrating that facial dynamics carry strong identity signatures. We introduce a drift-to-noise ratio (DNR) that quantifies the reliability of shape expression separation by measuring across-session shape changes relative to within-session variability. DNR strongly negatively correlates with recognition performance, confirming that unstable shape estimation compromises dynamic identification. Our findings reveal person-specific signatures in conversational facial dynamics, with implications for social perception and clinical assessment.

Anshul Gupta, Pierre Vuillecard, Arya Farkhondeh et al.

Contextual cues related to a person's pose and interactions with objects and other people in the scene can provide valuable information for gaze following. While existing methods have focused on dedicated cue extraction methods, in this work we investigate the zero-shot capabilities of Vision-Language Models (VLMs) for extracting a wide array of contextual cues to improve gaze following performance. We first evaluate various VLMs, prompting strategies, and in-context learning (ICL) techniques for zero-shot cue recognition performance. We then use these insights to extract contextual cues for gaze following, and investigate their impact when incorporated into a state of the art model for the task. Our analysis indicates that BLIP-2 is the overall top performing VLM and that ICL can improve performance. We also observe that VLMs are sensitive to the choice of the text prompt although ensembling over multiple text prompts can provide more robust performance. Additionally, we discover that using the entire image along with an ellipse drawn around the target person is the most effective strategy for visual prompting. For gaze following, incorporating the extracted cues results in better generalization performance, especially when considering a larger set of cues, highlighting the potential of this approach.

Angel Martínez-González, Michael Villamizar, Jean-Marc Odobez

We propose to leverage Transformer architectures for non-autoregressive human motion prediction. Our approach decodes elements in parallel from a query sequence, instead of conditioning on previous predictions such as instate-of-the-art RNN-based approaches. In such a way our approach is less computational intensive and potentially avoids error accumulation to long term elements in the sequence. In that context, our contributions are fourfold: (i) we frame human motion prediction as a sequence-to-sequence problem and propose a non-autoregressive Transformer to infer the sequences of poses in parallel; (ii) we propose to decode sequences of 3D poses from a query sequence generated in advance with elements from the input sequence;(iii) we propose to perform skeleton-based activity classification from the encoder memory, in the hope that identifying the activity can improve predictions;(iv) we show that despite its simplicity, our approach achieves competitive results in two public datasets, although surprisingly more for short term predictions rather than for long term ones.

Marco Ewerton, Angel Martínez-González, Jean-Marc Odobez

Humans effortlessly solve pushing tasks in everyday life but unlocking these capabilities remains a challenge in robotics because physics models of these tasks are often inaccurate or unattainable. State-of-the-art data-driven approaches learn to compensate for these inaccuracies or replace the approximated physics models altogether. Nevertheless, approaches like Deep Q-Networks (DQNs) suffer from local optima in large state-action spaces. Furthermore, they rely on well-chosen deep learning architectures and learning paradigms. In this paper, we propose to frame the learning of pushing policies (where to push and how) by DQNs as an image-to-image translation problem and exploit an Hourglass-based architecture. We present an architecture combining a predictor of which pushes lead to changes in the environment with a state-action value predictor dedicated to the pushing task. Moreover, we investigate positional information encoding to learn position-dependent policy behaviors. We demonstrate in simulation experiments with a UR5 robot arm that our overall architecture helps the DQN learn faster and achieve higher performance in a pushing task involving objects with unknown dynamics.

Angel Martínez-González, Michael Villamizar, Olivier Canévet et al.

We propose to leverage recent advances in reliable 2D pose estimation with Convolutional Neural Networks (CNN) to estimate the 3D pose of people from depth images in multi-person Human-Robot Interaction (HRI) scenarios. Our method is based on the observation that using the depth information to obtain 3D lifted points from 2D body landmark detections provides a rough estimate of the true 3D human pose, thus requiring only a refinement step. In that line our contributions are threefold. (i) we propose to perform 3D pose estimation from depth images by decoupling 2D pose estimation and 3D pose refinement; (ii) we propose a deep-learning approach that regresses the residual pose between the lifted 3D pose and the true 3D pose; (iii) we show that despite its simplicity, our approach achieves very competitive results both in accuracy and speed on two public datasets and is therefore appealing for multi-person HRI compared to recent state-of-the-art methods.

Angel Martínez-González, Michael Villamizar, Olivier Canévet et al.

Achieving robust multi-person 2D body landmark localization and pose estimation is essential for human behavior and interaction understanding as encountered for instance in HRI settings. Accurate methods have been proposed recently, but they usually rely on rather deep Convolutional Neural Network (CNN) architecture, thus requiring large computational and training resources. In this paper, we investigate different architectures and methodologies to address these issues and achieve fast and accurate multi-person 2D pose estimation. To foster speed, we propose to work with depth images, whose structure contains sufficient information about body landmarks while being simpler than textured color images and thus potentially requiring less complex CNNs for processing. In this context, we make the following contributions. i) we study several CNN architecture designs combining pose machines relying on the cascade of detectors concept with lightweight and efficient CNN structures; ii) to address the need for large training datasets with high variability, we rely on semi-synthetic data combining multi-person synthetic depth data with real sensor backgrounds; iii) we explore domain adaptation techniques to address the performance gap introduced by testing on real depth images; iv) to increase the accuracy of our fast lightweight CNN models, we investigate knowledge distillation at several architecture levels which effectively enhance performance. Experiments and results on synthetic and real data highlight the impact of our design choices, providing insights into methods addressing standard issues normally faced in practical applications, and resulting in architectures effectively matching our goal in both performance and speed.

Yu Yu, Jean-Marc Odobez

Although automatic gaze estimation is very important to a large variety of application areas, it is difficult to train accurate and robust gaze models, in great part due to the difficulty in collecting large and diverse data (annotating 3D gaze is expensive and existing datasets use different setups). To address this issue, our main contribution in this paper is to propose an effective approach to learn a low dimensional gaze representation without gaze annotations, which to the best of our best knowledge, is the first work to do so. The main idea is to rely on a gaze redirection network and use the gaze representation difference of the input and target images (of the redirection network) as the redirection variable. A redirection loss in image domain allows the joint training of both the redirection network and the gaze representation network. In addition, we propose a warping field regularization which not only provides an explicit physical meaning to the gaze representations but also avoids redirection distortions. Promising results on few-shot gaze estimation (competitive results can be achieved with as few as <= 100 calibration samples), cross-dataset gaze estimation, gaze network pretraining, and another task (head pose estimation) demonstrate the validity of our framework.

Angel Martínez-González, Michael Villamizar, Olivier Canévet et al.

We propose to combine recent Convolutional Neural Networks (CNN) models with depth imaging to obtain a reliable and fast multi-person pose estimation algorithm applicable to Human Robot Interaction (HRI) scenarios. Our hypothesis is that depth images contain less structures and are easier to process than RGB images while keeping the required information for human detection and pose inference, thus allowing the use of simpler networks for the task. Our contributions are threefold. (i) we propose a fast and efficient network based on residual blocks (called RPM) for body landmark localization from depth images; (ii) we created a public dataset DIH comprising more than 170k synthetic images of human bodies with various shapes and viewpoints as well as real (annotated) data for evaluation; (iii) we show that our model trained on synthetic data from scratch can perform well on real data, obtaining similar results to larger models initialized with pre-trained networks. It thus provides a good trade-off between performance and computation. Experiments on real data demonstrate the validity of our approach.

Mary Ellen Foster, Bart Craenen, Amol Deshmukh et al.

In the EU-funded MuMMER project, we have developed a social robot designed to interact naturally and flexibly with users in public spaces such as a shopping mall. We present the latest version of the robot system developed during the project. This system encompasses audio-visual sensing, social signal processing, conversational interaction, perspective taking, geometric reasoning, and motion planning. It successfully combines all these components in an overarching framework using the Robot Operating System (ROS) and has been deployed to a shopping mall in Finland interacting with customers. In this paper, we describe the system components, their interplay, and the resulting robot behaviours and scenarios provided at the shopping mall.

Yu Yu, Gang Liu, Jean-Marc Odobez

As an indicator of human attention gaze is a subtle behavioral cue which can be exploited in many applications. However, inferring 3D gaze direction is challenging even for deep neural networks given the lack of large amount of data (groundtruthing gaze is expensive and existing datasets use different setups) and the inherent presence of gaze biases due to person-specific difference. In this work, we address the problem of person-specific gaze model adaptation from only a few reference training samples. The main and novel idea is to improve gaze adaptation by generating additional training samples through the synthesis of gaze-redirected eye images from existing reference samples. In doing so, our contributions are threefold: (i) we design our gaze redirection framework from synthetic data, allowing us to benefit from aligned training sample pairs to predict accurate inverse mapping fields; (ii) we proposed a self-supervised approach for domain adaptation; (iii) we exploit the gaze redirection to improve the performance of person-specific gaze estimation. Extensive experiments on two public datasets demonstrate the validity of our gaze retargeting and gaze estimation framework.

Gang Liu, Yu Yu, Kenneth A. Funes Mora et al.

Non-invasive gaze estimation methods usually regress gaze directions directly from a single face or eye image. However, due to important variabilities in eye shapes and inner eye structures amongst individuals, universal models obtain limited accuracies and their output usually exhibit high variance as well as biases which are subject dependent. Therefore, increasing accuracy is usually done through calibration, allowing gaze predictions for a subject to be mapped to his/her actual gaze. In this paper, we introduce a novel image differential method for gaze estimation. We propose to directly train a differential convolutional neural network to predict the gaze differences between two eye input images of the same subject. Then, given a set of subject specific calibration images, we can use the inferred differences to predict the gaze direction of a novel eye sample. The assumption is that by allowing the comparison between two eye images, annoyance factors (alignment, eyelid closing, illumination perturbations) which usually plague single image prediction methods can be much reduced, allowing better prediction altogether. Experiments on 3 public datasets validate our approach which constantly outperforms state-of-the-art methods even when using only one calibration sample or when the latter methods are followed by subject specific gaze adaptation.

Nam Le, Jean-Marc Odobez

Collecting labeled data to train deep neural networks is costly and even impractical for many tasks. Thus, research effort has been focused in automatically curated datasets or unsupervised and weakly supervised learning. The common problem in these directions is learning with unreliable label information. In this paper, we address the tolerance of deep embedding learning losses against label noise, i.e. when the observed labels are different from the true labels. Specifically, we provide the sufficient conditions to achieve theoretical guarantees for the 2 common loss functions: marginal loss and triplet loss. From these theoretical results, we can estimate how sampling strategies and initialization can affect the level of resistance against label noise. The analysis also helps providing more effective guidelines in unsupervised and weakly supervised deep embedding learning.

Weipeng He, Petr Motlicek, Jean-Marc Odobez

We propose to use neural networks for simultaneous detection and localization of multiple sound sources in human-robot interaction. In contrast to conventional signal processing techniques, neural network-based sound source localization methods require fewer strong assumptions about the environment. Previous neural network-based methods have been focusing on localizing a single sound source, which do not extend to multiple sources in terms of detection and localization. In this paper, we thus propose a likelihood-based encoding of the network output, which naturally allows the detection of an arbitrary number of sources. In addition, we investigate the use of sub-band cross-correlation information as features for better localization in sound mixtures, as well as three different network architectures based on different motivations. Experiments on real data recorded from a robot show that our proposed methods significantly outperform the popular spatial spectrum-based approaches.

Nam Le, Jean-Marc Odobez

Learning speaker turn embeddings has shown considerable improvement in situations where conventional speaker modeling approaches fail. However, this improvement is relatively limited when compared to the gain observed in face embedding learning, which has been proven very successful for face verification and clustering tasks. Assuming that face and voices from the same identities share some latent properties (like age, gender, ethnicity), we propose three transfer learning approaches to leverage the knowledge from the face domain (learned from thousands of images and identities) for tasks in the speaker domain. These approaches, namely target embedding transfer, relative distance transfer, and clustering structure transfer, utilize the structure of the source face embedding space at different granularities to regularize the target speaker turn embedding space as optimizing terms. Our methods are evaluated on two public broadcast corpora and yield promising advances over competitive baselines in verification and audio clustering tasks, especially when dealing with short speaker utterances. The analysis of the results also gives insight into characteristics of the embedding spaces and shows their potential applications.