Christian Simon

Kaushik Roy, Christian Simon, Peyman Moghadam et al.

An ultimate objective in continual learning is to preserve knowledge learned in preceding tasks while learning new tasks. To mitigate forgetting prior knowledge, we propose a novel knowledge distillation technique that takes into the account the manifold structure of the latent/output space of a neural network in learning novel tasks. To achieve this, we propose to approximate the data manifold up-to its first order, hence benefiting from linear subspaces to model the structure and maintain the knowledge of a neural network while learning novel concepts. We demonstrate that the modeling with subspaces provides several intriguing properties, including robustness to noise and therefore effective for mitigating Catastrophic Forgetting in continual learning. We also discuss and show how our proposed method can be adopted to address both classification and segmentation problems. Empirically, we observe that our proposed method outperforms various continual learning methods on several challenging datasets including Pascal VOC, and Tiny-Imagenet. Furthermore, we show how the proposed method can be seamlessly combined with existing learning approaches to improve their performances. The codes of this article will be available at https://github.com/csiro-robotics/SDCL.

Christian Simon, Masoud Faraki, Yi-Hsuan Tsai et al.

Humans have the ability to accumulate knowledge of new tasks in varying conditions, but deep neural networks often suffer from catastrophic forgetting of previously learned knowledge after learning a new task. Many recent methods focus on preventing catastrophic forgetting under the assumption of train and test data following similar distributions. In this work, we consider a more realistic scenario of continual learning under domain shifts where the model must generalize its inference to an unseen domain. To this end, we encourage learning semantically meaningful features by equipping the classifier with class similarity metrics as learning parameters which are obtained through Mahalanobis similarity computations. Learning of the backbone representation along with these extra parameters is done seamlessly in an end-to-end manner. In addition, we propose an approach based on the exponential moving average of the parameters for better knowledge distillation. We demonstrate that, to a great extent, existing continual learning algorithms fail to handle the forgetting issue under multiple distributions, while our proposed approach learns new tasks under domain shift with accuracy boosts up to 10% on challenging datasets such as DomainNet and OfficeHome.

Yuren Cong, Mengmeng Xu, Christian Simon et al.

Text-to-video editing aims to edit the visual appearance of a source video conditional on textual prompts. A major challenge in this task is to ensure that all frames in the edited video are visually consistent. Most recent works apply advanced text-to-image diffusion models to this task by inflating 2D spatial attention in the U-Net into spatio-temporal attention. Although temporal context can be added through spatio-temporal attention, it may introduce some irrelevant information for each patch and therefore cause inconsistency in the edited video. In this paper, for the first time, we introduce optical flow into the attention module in the diffusion model's U-Net to address the inconsistency issue for text-to-video editing. Our method, FLATTEN, enforces the patches on the same flow path across different frames to attend to each other in the attention module, thus improving the visual consistency in the edited videos. Additionally, our method is training-free and can be seamlessly integrated into any diffusion-based text-to-video editing methods and improve their visual consistency. Experiment results on existing text-to-video editing benchmarks show that our proposed method achieves the new state-of-the-art performance. In particular, our method excels in maintaining the visual consistency in the edited videos.

Abdelmalek Mouazer, Sophie Dubois, Romain Léguillon et al.

Background: Medication review is a structured interview of the patient, performed by the pharmacist and aimed at optimizing drug treatments. In practice, medication review is a long and cognitively-demanding task that requires specific knowledge. Clinical practice guidelines have been proposed, but their application is tedious. Methods: We designed ABiMed, a clinical decision support system for medication reviews, based on the implementation of the STOPP/START v2 guidelines and on the visual presentation of aggregated drug knowledge using tables, graphs and flower glyphs. We evaluated ABiMed with 39 community pharmacists during a randomized simulation trial, each pharmacist performing a medication review for two fictitious patients without ABiMed, and two others with ABiMed. We recorded the problems identified by the pharmacists, the interventions proposed, the response time, the perceived usability and the comments. Pharmacists' medication reviews were compared to an expert-designed gold standard. Results: With ABiMed, pharmacists found 1.6 times more relevant drug-related problems during the medication review (p=1.1e-12) and proposed better interventions (p=9.8e-9), without needing more time (p=0.56). The System Usability Scale score is 82.7, which is ranked "excellent". In their comments, pharmacists appreciated the visual aspect of ABiMed and its ability to compare the current treatment with the proposed one. A multifactor analysis showed no difference in the support offered by ABiMed according to the pharmacist's age or sex, in terms of percentage of problems identified or quality of the proposed interventions. Conclusions: The use of an intelligent and visual clinical decision support system can help pharmacists when they perform medication reviews. Our main perspective is the validation of the system in clinical conditions.

Christian Simon, Masato Ishii, Wei-Yao Wang et al.

Scaling multimodal alignment between video and audio is challenging, particularly due to limited data and the mismatch between text descriptions and frame-level video information. In this work, we tackle the scaling challenge in multimodal-to-audio generation, examining whether models trained on short instances can generalize to longer ones during testing. To tackle this challenge, we present multimodal hierarchical networks so-called MMHNet, an enhanced extension of state-of-the-art video-to-audio models. Our approach integrates a hierarchical method and non-causal Mamba to support long-form audio generation. Our proposed method significantly improves long audio generation up to more than 5 minutes. We also prove that training short and testing long is possible in the video-to-audio generation tasks without training on the longer durations. We show in our experiments that our proposed method could achieve remarkable results on long-video to audio benchmarks, beating prior works in video-to-audio tasks. Moreover, we showcase our model capability in generating more than 5 minutes, while prior video-to-audio methods fall short in generating with long durations.

Samuel Cahyawijaya, Holy Lovenia, Joel Ruben Antony Moniz et al. · cambridge

Southeast Asia (SEA) is a region of extraordinary linguistic and cultural diversity, yet it remains significantly underrepresented in vision-language (VL) research. This often results in artificial intelligence (AI) models that fail to capture SEA cultural nuances. To fill this gap, we present SEA-VL, an open-source initiative dedicated to developing high-quality, culturally relevant data for SEA languages. By involving contributors from SEA countries, SEA-VL aims to ensure better cultural relevance and diversity, fostering greater inclusivity of underrepresented languages in VL research. Beyond crowdsourcing, our initiative goes one step further in the exploration of the automatic collection of culturally relevant images through crawling and image generation. First, we find that image crawling achieves approximately ~85% cultural relevance while being more cost- and time-efficient than crowdsourcing. Second, despite the substantial progress in generative vision models, synthetic images remain unreliable in accurately reflecting SEA cultures. The generated images often fail to reflect the nuanced traditions and cultural contexts of the region. Collectively, we gather 1.28M SEA culturally-relevant images, more than 50 times larger than other existing datasets. Through SEA-VL, we aim to bridge the representation gap in SEA, fostering the development of more inclusive AI systems that authentically represent diverse cultures across SEA.

Joshua Santoso, Christian Simon, Williem

Diffusion models have gained attention for image editing yielding impressive results in text-to-image tasks. On the downside, one might notice that generated images of stable diffusion models suffer from deteriorated details. This pitfall impacts image editing tasks that require information preservation e.g., scene text editing. As a desired result, the model must show the capability to replace the text on the source image to the target text while preserving the details e.g., color, font size, and background. To leverage the potential of diffusion models, in this work, we introduce Diffusion-BasEd Scene Text manipulation Network so-called DBEST. Specifically, we design two adaptation strategies, namely one-shot style adaptation and text-recognition guidance. In experiments, we thoroughly assess and compare our proposed method against state-of-the-arts on various scene text datasets, then provide extensive ablation studies for each granularity to analyze our performance gain. Also, we demonstrate the effectiveness of our proposed method to synthesize scene text indicated by competitive Optical Character Recognition (OCR) accuracy. Our method achieves 94.15% and 98.12% on COCO-text and ICDAR2013 datasets for character-level evaluation.

Shuyang Cui, Zhi Zhong, Qiyu Wu et al.

Current methods for creating drum loop audio in digital music production, such as using one-shot samples or resampling, often demand non-trivial efforts of creators. While recent generative models achieve high fidelity and adhere to text, they lack the specific control needed for such a task. Existing symbolic-to-audio research often focuses on single, tonal instruments, leaving the challenge of polyphonic, percussive drum synthesis unaddressed. We address this gap by introducing ``Break-the-Beat!,'' a model capable of rendering a drum MIDI with the timbre of a reference audio. It is built by fine-tuning a pre-trained text-to-audio model with our proposed content encoder and a effective hybrid conditioning mechanism. To enable this, we construct a new dataset of paired target-reference drum audio from existing drum audio datasets. Experiments demonstrate that our model generates high-quality drum audio that follows high-resolution drum MIDI, achieving strong performance across metrics of audio quality, rhythmic alignment, and beat continuity. This offer producers a new, controllable tool for creative production. Demo page: https://ik4sumii.github.io/break-the-beat/

Yuanhong Chen, Kazuki Shimada, Christian Simon et al.

Binaural audio generation (BAG) aims to convert monaural audio to stereo audio using visual prompts, requiring a deep understanding of spatial and semantic information. However, current models risk overfitting to room environments and lose fine-grained spatial details. In this paper, we propose a new audio-visual binaural generation model incorporating an audio-visual conditional normalisation layer that dynamically aligns the mean and variance of the target difference audio features using visual context, along with a new contrastive learning method to enhance spatial sensitivity by mining negative samples from shuffled visual features. We also introduce a cost-efficient way to utilise test-time augmentation in video data to enhance performance. Our approach achieves state-of-the-art generation accuracy on the FAIR-Play and MUSIC-Stereo benchmarks.

Abdelmalek Mouazer, Romain Léguillon, Nada Boudegzdame et al.

Background: Polypharmacy, i.e. taking five drugs or more, is both a public health and an economic issue. Medication reviews are structured interviews of the patient by the community pharmacist, aiming at optimizing the drug treatment and deprescribing useless, redundant or dangerous drugs. However, they remain difficult to perform and time-consuming. Several clinical decision support systems were developed for helping clinicians to manage polypharmacy. However, most were limited to the implementation of clinical practice guidelines. In this work, our objective is to design an innovative clinical decision support system for medication reviews and polypharmacy management, named ABiMed. Methods: ABiMed associates several approaches: guidelines implementation, but the automatic extraction of patient data from the GP's electronic health record and its transfer to the pharmacist, and the visual presentation of contextualized drug knowledge using visual analytics. We performed an ergonomic assessment and qualitative evaluations involving pharmacists and GPs during focus groups and workshops. Results: We describe the proposed architecture, which allows a collaborative multi-user usage. We present the various screens of ABiMed for entering or verifying patient data, for accessing drug knowledge (posology, adverse effects, interactions), for viewing STOPP/START rules and for suggesting modification to the treatment. Qualitative evaluations showed that health professionals were highly interested by our approach, associating the automatic guidelines execution with the visual presentation of drug knowledge. Conclusions: The association of guidelines implementation with visual presentation of knowledge is a promising approach for managing polypharmacy. Future works will focus on the improvement and the evaluation of ABiMed.

Christian Simon, Sen He, Juan-Manuel Perez-Rua et al.

Solving image-to-3D from a single view is an ill-posed problem, and current neural reconstruction methods addressing it through diffusion models still rely on scene-specific optimization, constraining their generalization capability. To overcome the limitations of existing approaches regarding generalization and consistency, we introduce a novel neural rendering technique. Our approach employs the signed distance function as the surface representation and incorporates generalizable priors through geometry-encoding volumes and HyperNetworks. Specifically, our method builds neural encoding volumes from generated multi-view inputs. We adjust the weights of the SDF network conditioned on an input image at test-time to allow model adaptation to novel scenes in a feed-forward manner via HyperNetworks. To mitigate artifacts derived from the synthesized views, we propose the use of a volume transformer module to improve the aggregation of image features instead of processing each viewpoint separately. Through our proposed method, dubbed as Hyper-VolTran, we avoid the bottleneck of scene-specific optimization and maintain consistency across the images generated from multiple viewpoints. Our experiments show the advantages of our proposed approach with consistent results and rapid generation.

Leila Mahmoodi, Peyman Moghadam, Munawar Hayat et al.

We introduce Flashback Learning (FL), a novel method designed to harmonize the stability and plasticity of models in Continual Learning (CL). Unlike prior approaches that primarily focus on regularizing model updates to preserve old information while learning new concepts, FL explicitly balances this trade-off through a bidirectional form of regularization. This approach effectively guides the model to swiftly incorporate new knowledge while actively retaining its old knowledge. FL operates through a two-phase training process and can be seamlessly integrated into various CL methods, including replay, parameter regularization, distillation, and dynamic architecture techniques. In designing FL, we use two distinct knowledge bases: one to enhance plasticity and another to improve stability. FL ensures a more balanced model by utilizing both knowledge bases to regularize model updates. Theoretically, we analyze how the FL mechanism enhances the stability-plasticity balance. Empirically, FL demonstrates tangible improvements over baseline methods within the same training budget. By integrating FL into at least one representative baseline from each CL category, we observed an average accuracy improvement of up to 4.91% in Class-Incremental and 3.51% in Task-Incremental settings on standard image classification benchmarks. Additionally, measurements of the stability-to-plasticity ratio confirm that FL effectively enhances this balance. FL also outperforms state-of-the-art CL methods on more challenging datasets like ImageNet.

Koichi Saito, Julian Tanke, Christian Simon et al.

Prevailing Video-to-Audio (V2A) generation models operate offline, assuming an entire video sequence or chunks of frames are available beforehand. This critically limits their use in interactive applications such as live content creation and emerging generative world models. To address this gap, we introduce the novel task of frame-level online V2A generation, where a model autoregressively generates audio from video without access to future video frames. Furthermore, we propose SoundReactor, which, to the best of our knowledge, is the first simple yet effective framework explicitly tailored for this task. Our design enforces end-to-end causality and targets low per-frame latency with audio-visual synchronization. Our model's backbone is a decoder-only causal transformer over continuous audio latents. For vision conditioning, it leverages grid (patch) features extracted from the smallest variant of the DINOv2 vision encoder, which are aggregated into a single token per frame to maintain end-to-end causality and efficiency. The model is trained through a diffusion pre-training followed by consistency fine-tuning to accelerate the diffusion head decoding. On a benchmark of diverse gameplay videos from AAA titles, our model successfully generates semantically and temporally aligned, high-quality full-band stereo audio, validated by both objective and human evaluations. Furthermore, our model achieves low per-frame waveform-level latency (26.3ms with the head NFE=1, 31.5ms with NFE=4) on 30FPS, 480p videos using a single H100. Demo samples are available at https://koichi-saito-sony.github.io/soundreactor/.

Christian Simon, Masato Ishii, Akio Hayakawa et al.

In the recent development of conditional diffusion models still require heavy supervised fine-tuning for performing control on a category of tasks. Training-free conditioning via guidance with off-the-shelf models is a favorable alternative to avoid further fine-tuning on the base model. However, the existing training-free guidance frameworks either have heavy memory requirements or offer sub-optimal control due to rough estimation. These shortcomings limit the applicability to control diffusion models that require intense computation, such as Text-to-Video (T2V) diffusion models. In this work, we propose Taming Inference Time Alignment for Guided Text-to-Video Diffusion Model, so-called TITAN-Guide, which overcomes memory space issues, and provides more optimal control in the guidance process compared to the counterparts. In particular, we develop an efficient method for optimizing diffusion latents without backpropagation from a discriminative guiding model. In particular, we study forward gradient descents for guided diffusion tasks with various options on directional directives. In our experiments, we demonstrate the effectiveness of our approach in efficiently managing memory during latent optimization, while previous methods fall short. Our proposed approach not only minimizes memory requirements but also significantly enhances T2V performance across a range of diffusion guidance benchmarks. Code, models, and demo are available at https://titanguide.github.io.

Christian Simon, Piotr Koniusz, Mehrtash Harandi

Even with the luxury of having abundant data, multi-label classification is widely known to be a challenging task to address. This work targets the problem of multi-label meta-learning, where a model learns to predict multiple labels within a query (e.g., an image) by just observing a few supporting examples. In doing so, we first propose a benchmark for Few-Shot Learning (FSL) with multiple labels per sample. Next, we discuss and extend several solutions specifically designed to address the conventional and single-label FSL, to work in the multi-label regime. Lastly, we introduce a neural module to estimate the label count of a given sample by exploiting the relational inference. We will show empirically the benefit of the label count module, the label propagation algorithm, and the extensions of conventional FSL methods on three challenging datasets, namely MS-COCO, iMaterialist, and Open MIC. Overall, our thorough experiments suggest that the proposed label-propagation algorithm in conjunction with the neural label count module (NLC) shall be considered as the method of choice.

Christian Simon, Piotr Koniusz, Lars Petersson et al.

Neural Architecture Search (NAS) is the game changer in designing robust neural architectures. Architectures designed by NAS outperform or compete with the best manual network designs in terms of accuracy, size, memory footprint and FLOPs. That said, previous studies focus on developing NAS algorithms for clean high quality data, a restrictive and somewhat unrealistic assumption. In this paper, focusing on the differentiable NAS algorithms, we show that vanilla NAS algorithms suffer from a performance loss if class labels are noisy. To combat this issue, we make use of the principle of information bottleneck as a regularizer. This leads us to develop a noise injecting operation that is included during the learning process, preventing the network from learning from noisy samples. Our empirical evaluations show that the noise injecting operation does not degrade the performance of the NAS algorithm if the data is indeed clean. In contrast, if the data is noisy, the architecture learned by our algorithm comfortably outperforms algorithms specifically equipped with sophisticated mechanisms to learn in the presence of label noise. In contrast to many algorithms designed to work in the presence of noisy labels, prior knowledge about the properties of the noise and its characteristics are not required for our algorithm.

Christian Simon, Piotr Koniusz, Mehrtash Harandi

Neural networks notoriously suffer from the problem of catastrophic forgetting, the phenomenon of forgetting the past knowledge when acquiring new knowledge. Overcoming catastrophic forgetting is of significant importance to emulate the process of "incremental learning", where the model is capable of learning from sequential experience in an efficient and robust way. State-of-the-art techniques for incremental learning make use of knowledge distillation towards preventing catastrophic forgetting. Therein, one updates the network while ensuring that the network's responses to previously seen concepts remain stable throughout updates. This in practice is done by minimizing the dissimilarity between current and previous responses of the network one way or another. Our work contributes a novel method to the arsenal of distillation techniques. In contrast to the previous state of the art, we propose to firstly construct low-dimensional manifolds for previous and current responses and minimize the dissimilarity between the responses along the geodesic connecting the manifolds. This induces a more formidable knowledge distillation with smooth properties which preserves the past knowledge more efficiently as observed by our comprehensive empirical study.

Jie Hong, Pengfei Fang, Weihao Li et al.

Few-shot learning aims to correctly recognize query samples from unseen classes given a limited number of support samples, often by relying on global embeddings of images. In this paper, we propose to equip the backbone network with an attention agent, which is trained by reinforcement learning. The policy gradient algorithm is employed to train the agent towards adaptively localizing the representative regions on feature maps over time. We further design a reward function based on the prediction of the held-out data, thus helping the attention mechanism to generalize better across the unseen classes. The extensive experiments show, with the help of the reinforced attention, that our embedding network has the capability to progressively generate a more discriminative representation in few-shot learning. Moreover, experiments on the task of image classification also show the effectiveness of the proposed design.

Christian Simon, Matteo Torcoli, Jouni Paulus

Broadcasting and streaming services still suffer from various levels of accessibility barriers for a significant portion of the population, limiting the access to information and culture, and in the most severe cases limiting the empowerment of people. This paper provides a brief overview of some of the most common accessibility barriers encountered. It then gives a short introduction to object-based audio (OBA) production and transport, focusing on the aspects relevant for lowering accessibility barriers. MPEG-H Audio is used as a concrete example of an OBA system already deployed. Two example cases (dialog enhancement and audio description) are used to demonstrate in detail the simplicity of producing MPEG-H Audio content providing improved accessibility. Several other possibilities are outlined briefly. We show that using OBA for broadcasting and streaming content allows offering several accessibility features in a flexible manner, requiring only small changes to the existing production workflow, assuming the receiver supports the functionality.

Jean-Baptiste Lamy, Vahid Ebrahiminia, Brigitte Seroussi et al.

Clinical decision support systems have been developed to help physicians to take clinical guidelines into account during consultations. The ASTI critiquing module is one such systems; it provides the physician with automatic criticisms when a drug prescription does not follow the guidelines. It was initially developed for hypertension and type 2 diabetes, but is designed to be generic enough for application to all chronic diseases. We present here the results of usability and satisfaction evaluations for the ASTI critiquing module, obtained with GPs for a newly implemented guideline concerning dyslipaemia, and we discuss the lessons learnt and the difficulties encountered when building a generic DSS for critiquing physicians' prescriptions.