This paper reviews the Challenge on Video Saliency Prediction at AIM 2024. The goal of the participants was to develop a method for predicting accurate saliency maps for the provided set of video sequences. Saliency maps are widely exploited in various applications, including video compression, quality assessment, visual perception studies, the advertising industry, etc. For this competition, a previously unused large-scale audio-visual mouse saliency (AViMoS) dataset of 1500 videos with more than 70 observers per video was collected using crowdsourced mouse tracking. The dataset collection methodology has been validated using conventional eye-tracking data and has shown high consistency. Over 30 teams registered in the challenge, and there are 7 teams that submitted the results in the final phase. The final phase solutions were tested and ranked by commonly used quality metrics on a private test subset. The results of this evaluation and the descriptions of the solutions are presented in this report. All data, including the private test subset, is made publicly available on the challenge homepage - https://challenges.videoprocessing.ai/challenges/video-saliency-prediction.html.
With the rapid development of depth sensor, more and more RGB-D videos could be obtained. Identifying the foreground in RGB-D videos is a fundamental and important task. However, the existing salient object detection (SOD) works only focus on either static RGB-D images or RGB videos, ignoring the collaborating of RGB-D and video information. In this paper, we first collect a new annotated RGB-D video SOD (ViDSOD-100) dataset, which contains 100 videos within a total of 9,362 frames, acquired from diverse natural scenes. All the frames in each video are manually annotated to a high-quality saliency annotation. Moreover, we propose a new baseline model, named attentive triple-fusion network (ATF-Net), for RGB-D video salient object detection. Our method aggregates the appearance information from an input RGB image, spatio-temporal information from an estimated motion map, and the geometry information from the depth map by devising three modality-specific branches and a multi-modality integration branch. The modality-specific branches extract the representation of different inputs, while the multi-modality integration branch combines the multi-level modality-specific features by introducing the encoder feature aggregation (MEA) modules and decoder feature aggregation (MDA) modules. The experimental findings conducted on both our newly introduced ViDSOD-100 dataset and the well-established DAVSOD dataset highlight the superior performance of the proposed ATF-Net. This performance enhancement is demonstrated both quantitatively and qualitatively, surpassing the capabilities of current state-of-the-art techniques across various domains, including RGB-D saliency detection, video saliency detection, and video object segmentation. Our data and our code are available at github.com/jhl-Det/RGBD_Video_SOD.
In this work, we present an integrated system for spatiotemporal summarization of 360-degrees videos. The video summary production mainly involves the detection of salient events and their synopsis into a concise summary. The analysis relies on state-of-the-art methods for saliency detection in 360-degrees video (ATSal and SST-Sal) and video summarization (CA-SUM). It also contains a mechanism that classifies a 360-degrees video based on the use of static or moving camera during recording and decides which saliency detection method will be used, as well as a 2D video production component that is responsible to create a conventional 2D video containing the salient events in the 360-degrees video. Quantitative evaluations using two datasets for 360-degrees video saliency detection (VR-EyeTracking, Sports-360) show the accuracy and positive impact of the developed decision mechanism, and justify our choice to use two different methods for detecting the salient events. A qualitative analysis using content from these datasets, gives further insights about the functionality of the decision mechanism, shows the pros and cons of each used saliency detection method and demonstrates the advanced performance of the trained summarization method against a more conventional approach.
360$^\circ$ video saliency detection is one of the challenging benchmarks for 360$^\circ$ video understanding since non-negligible distortion and discontinuity occur in the projection of any format of 360$^\circ$ videos, and capture-worthy viewpoint in the omnidirectional sphere is ambiguous by nature. We present a new framework named Panoramic Vision Transformer (PAVER). We design the encoder using Vision Transformer with deformable convolution, which enables us not only to plug pretrained models from normal videos into our architecture without additional modules or finetuning but also to perform geometric approximation only once, unlike previous deep CNN-based approaches. Thanks to its powerful encoder, PAVER can learn the saliency from three simple relative relations among local patch features, outperforming state-of-the-art models for the Wild360 benchmark by large margins without supervision or auxiliary information like class activation. We demonstrate the utility of our saliency prediction model with the omnidirectional video quality assessment task in VQA-ODV, where we consistently improve performance without any form of supervision, including head movement.
Video saliency detection (VSD) aims at fast locating the most attractive objects/things/patterns in a given video clip. Existing VSD-related works have mainly relied on the visual system but paid less attention to the audio aspect, while, actually, our audio system is the most vital complementary part to our visual system. Also, audio-visual saliency detection (AVSD), one of the most representative research topics for mimicking human perceptual mechanisms, is currently in its infancy, and none of the existing survey papers have touched on it, especially from the perspective of saliency detection. Thus, the ultimate goal of this paper is to provide an extensive review to bridge the gap between audio-visual fusion and saliency detection. In addition, as another highlight of this review, we have provided a deep insight into key factors which could directly determine the performances of AVSD deep models, and we claim that the audio-visual consistency degree (AVC) -- a long-overlooked issue, can directly influence the effectiveness of using audio to benefit its visual counterpart when performing saliency detection. Moreover, in order to make the AVC issue more practical and valuable for future followers, we have newly equipped almost all existing publicly available AVSD datasets with additional frame-wise AVC labels. Based on these upgraded datasets, we have conducted extensive quantitative evaluations to ground our claim on the importance of AVC in the AVSD task. In a word, both our ideas and new sets serve as a convenient platform with preliminaries and guidelines, all of which are very potential to facilitate future works in promoting state-of-the-art (SOTA) performance further.
Saliency prediction refers to the computational task of modeling overt attention. Social cues greatly influence our attention, consequently altering our eye movements and behavior. To emphasize the efficacy of such features, we present a neural model for integrating social cues and weighting their influences. Our model consists of two stages. During the first stage, we detect two social cues by following gaze, estimating gaze direction, and recognizing affect. These features are then transformed into spatiotemporal maps through image processing operations. The transformed representations are propagated to the second stage (GASP) where we explore various techniques of late fusion for integrating social cues and introduce two sub-networks for directing attention to relevant stimuli. Our experiments indicate that fusion approaches achieve better results for static integration methods, whereas non-fusion approaches for which the influence of each modality is unknown, result in better outcomes when coupled with recurrent models for dynamic saliency prediction. We show that gaze direction and affective representations contribute a prediction to ground-truth correspondence improvement of at least 5% compared to dynamic saliency models without social cues. Furthermore, affective representations improve GASP, supporting the necessity of considering affect-biased attention in predicting saliency.
Thanks to the rapid advances in deep learning techniques and the wide availability of large-scale training sets, the performance of video saliency detection models has been improving steadily and significantly. However, deep learning-based visualaudio fixation prediction is still in its infancy. At present, only a few visual-audio sequences have been furnished, with real fixations being recorded in real visual-audio environments. Hence, it would be neither efficient nor necessary to recollect real fixations under the same visual-audio circumstances. To address this problem, this paper promotes a novel approach in a weakly supervised manner to alleviate the demand of large-scale training sets for visual-audio model training. By using only the video category tags, we propose the selective class activation mapping (SCAM) and its upgrade (SCAM+). In the spatial-temporal-audio circumstance, the former follows a coarse-to-fine strategy to select the most discriminative regions, and these regions are usually capable of exhibiting high consistency with the real human-eye fixations. The latter equips the SCAM with an additional multi-granularity perception mechanism, making the whole process more consistent with that of the real human visual system. Moreover, we distill knowledge from these regions to obtain complete new spatial-temporal-audio (STA) fixation prediction (FP) networks, enabling broad applications in cases where video tags are not available. Without resorting to any real human-eye fixation, the performances of these STA FP networks are comparable to those of fully supervised networks. The code and results are publicly available at https://github.com/guotaowang/STANet.
Significant performance improvement has been achieved for fully-supervised video salient object detection with the pixel-wise labeled training datasets, which are time-consuming and expensive to obtain. To relieve the burden of data annotation, we present the first weakly supervised video salient object detection model based on relabeled "fixation guided scribble annotations". Specifically, an "Appearance-motion fusion module" and bidirectional ConvLSTM based framework are proposed to achieve effective multi-modal learning and long-term temporal context modeling based on our new weak annotations. Further, we design a novel foreground-background similarity loss to further explore the labeling similarity across frames. A weak annotation boosting strategy is also introduced to boost our model performance with a new pseudo-label generation technique. Extensive experimental results on six benchmark video saliency detection datasets illustrate the effectiveness of our solution.
We propose the ViNet architecture for audio-visual saliency prediction. ViNet is a fully convolutional encoder-decoder architecture. The encoder uses visual features from a network trained for action recognition, and the decoder infers a saliency map via trilinear interpolation and 3D convolutions, combining features from multiple hierarchies. The overall architecture of ViNet is conceptually simple; it is causal and runs in real-time (60 fps). ViNet does not use audio as input and still outperforms the state-of-the-art audio-visual saliency prediction models on nine different datasets (three visual-only and six audio-visual datasets). ViNet also surpasses human performance on the CC, SIM and AUC metrics for the AVE dataset, and to our knowledge, it is the first network to do so. We also explore a variation of ViNet architecture by augmenting audio features into the decoder. To our surprise, upon sufficient training, the network becomes agnostic to the input audio and provides the same output irrespective of the input. Interestingly, we also observe similar behaviour in the previous state-of-the-art models \cite{tsiami2020stavis} for audio-visual saliency prediction. Our findings contrast with previous works on deep learning-based audio-visual saliency prediction, suggesting a clear avenue for future explorations incorporating audio in a more effective manner. The code and pre-trained models are available at https://github.com/samyak0210/ViNet.
In this work, we propose a 3D fully convolutional architecture for video saliency prediction that employs hierarchical supervision on intermediate maps (referred to as conspicuity maps) generated using features extracted at different abstraction levels. We provide the base hierarchical learning mechanism with two techniques for domain adaptation and domain-specific learning. For the former, we encourage the model to unsupervisedly learn hierarchical general features using gradient reversal at multiple scales, to enhance generalization capabilities on datasets for which no annotations are provided during training. As for domain specialization, we employ domain-specific operations (namely, priors, smoothing and batch normalization) by specializing the learned features on individual datasets in order to maximize performance. The results of our experiments show that the proposed model yields state-of-the-art accuracy on supervised saliency prediction. When the base hierarchical model is empowered with domain-specific modules, performance improves, outperforming state-of-the-art models on three out of five metrics on the DHF1K benchmark and reaching the second-best results on the other two. When, instead, we test it in an unsupervised domain adaptation setting, by enabling hierarchical gradient reversal layers, we obtain performance comparable to supervised state-of-the-art.
With the rapid development of deep learning techniques, image saliency deep models trained solely by spatial information have occasionally achieved detection performance for video data comparable to that of the models trained by both spatial and temporal information. However, due to the lesser consideration of temporal information, the image saliency deep models may become fragile in the video sequences dominated by temporal information. Thus, the most recent video saliency detection approaches have adopted the network architecture starting with a spatial deep model that is followed by an elaborately designed temporal deep model. However, such methods easily encounter the performance bottleneck arising from the single stream learning methodology, so the overall detection performance is largely determined by the spatial deep model. In sharp contrast to the current mainstream methods, this paper proposes a novel plug-and-play scheme to weakly retrain a pretrained image saliency deep model for video data by using the newly sensed and coded temporal information. Thus, the retrained image saliency deep model will be able to maintain temporal saliency awareness, achieving much improved detection performance. Moreover, our method is simple yet effective for adapting any off-the-shelf pre-trained image saliency deep model to obtain high-quality video saliency detection. Additionally, both the data and source code of our method are publicly available.
The current main stream methods formulate their video saliency mainly from two independent venues, i.e., the spatial and temporal branches. As a complementary component, the main task for the temporal branch is to intermittently focus the spatial branch on those regions with salient movements. In this way, even though the overall video saliency quality is heavily dependent on its spatial branch, however, the performance of the temporal branch still matter. Thus, the key factor to improve the overall video saliency is how to further boost the performance of these branches efficiently. In this paper, we propose a novel spatiotemporal network to achieve such improvement in a full interactive fashion. We integrate a lightweight temporal model into the spatial branch to coarsely locate those spatially salient regions which are correlated with trustworthy salient movements. Meanwhile, the spatial branch itself is able to recurrently refine the temporal model in a multi-scale manner. In this way, both the spatial and temporal branches are able to interact with each other, achieving the mutual performance improvement. Our method is easy to implement yet effective, achieving high quality video saliency detection in real-time speed with 50 FPS.
Visual saliency modeling for images and videos is treated as two independent tasks in recent computer vision literature. While image saliency modeling is a well-studied problem and progress on benchmarks like SALICON and MIT300 is slowing, video saliency models have shown rapid gains on the recent DHF1K benchmark. Here, we take a step back and ask: Can image and video saliency modeling be approached via a unified model, with mutual benefit? We identify different sources of domain shift between image and video saliency data and between different video saliency datasets as a key challenge for effective joint modelling. To address this we propose four novel domain adaptation techniques - Domain-Adaptive Priors, Domain-Adaptive Fusion, Domain-Adaptive Smoothing and Bypass-RNN - in addition to an improved formulation of learned Gaussian priors. We integrate these techniques into a simple and lightweight encoder-RNN-decoder-style network, UNISAL, and train it jointly with image and video saliency data. We evaluate our method on the video saliency datasets DHF1K, Hollywood-2 and UCF-Sports, and the image saliency datasets SALICON and MIT300. With one set of parameters, UNISAL achieves state-of-the-art performance on all video saliency datasets and is on par with the state-of-the-art for image saliency datasets, despite faster runtime and a 5 to 20-fold smaller model size compared to all competing deep methods. We provide retrospective analyses and ablation studies which confirm the importance of the domain shift modeling. The code is available at https://github.com/rdroste/unisal
Due to a variety of motions across different frames, it is highly challenging to learn an effective spatiotemporal representation for accurate video saliency prediction (VSP). To address this issue, we develop an effective spatiotemporal feature alignment network tailored to VSP, mainly including two key sub-networks: a multi-scale deformable convolutional alignment network (MDAN) and a bidirectional convolutional Long Short-Term Memory (Bi-ConvLSTM) network. The MDAN learns to align the features of the neighboring frames to the reference one in a coarse-to-fine manner, which can well handle various motions. Specifically, the MDAN owns a pyramidal feature hierarchy structure that first leverages deformable convolution (Dconv) to align the lower-resolution features across frames, and then aggregates the aligned features to align the higher-resolution features, progressively enhancing the features from top to bottom. The output of MDAN is then fed into the Bi-ConvLSTM for further enhancement, which captures the useful long-time temporal information along forward and backward timing directions to effectively guide attention orientation shift prediction under complex scene transformation. Finally, the enhanced features are decoded to generate the predicted saliency map. The proposed model is trained end-to-end without any intricate post processing. Extensive evaluations on four VSP benchmark datasets demonstrate that the proposed method achieves favorable performance against state-of-the-art methods. The source codes and all the results will be released.
TASED-Net is a 3D fully-convolutional network architecture for video saliency detection. It consists of two building blocks: first, the encoder network extracts low-resolution spatiotemporal features from an input clip of several consecutive frames, and then the following prediction network decodes the encoded features spatially while aggregating all the temporal information. As a result, a single prediction map is produced from an input clip of multiple frames. Frame-wise saliency maps can be predicted by applying TASED-Net in a sliding-window fashion to a video. The proposed approach assumes that the saliency map of any frame can be predicted by considering a limited number of past frames. The results of our extensive experiments on video saliency detection validate this assumption and demonstrate that our fully-convolutional model with temporal aggregation method is effective. TASED-Net significantly outperforms previous state-of-the-art approaches on all three major large-scale datasets of video saliency detection: DHF1K, Hollywood2, and UCFSports. After analyzing the results qualitatively, we observe that our model is especially better at attending to salient moving objects.
This paper investigates modifying an existing neural network architecture for static saliency prediction using two types of recurrences that integrate information from the temporal domain. The first modification is the addition of a ConvLSTM within the architecture, while the second is a conceptually simple exponential moving average of an internal convolutional state. We use weights pre-trained on the SALICON dataset and fine-tune our model on DHF1K. Our results show that both modifications achieve state-of-the-art results and produce similar saliency maps. Source code is available at https://git.io/fjPiB.
Recently, with the advent of deep convolutional neural networks (DCNN), the improvements in visual saliency prediction research are impressive. One possible direction to approach the next improvement is to fully characterize the multi-scale saliency-influential factors with a computationally-friendly module in DCNN architectures. In this work, we proposed an end-to-end dilated inception network (DINet) for visual saliency prediction. It captures multi-scale contextual features effectively with very limited extra parameters. Instead of utilizing parallel standard convolutions with different kernel sizes as the existing inception module, our proposed dilated inception module (DIM) uses parallel dilated convolutions with different dilation rates which can significantly reduce the computation load while enriching the diversity of receptive fields in feature maps. Moreover, the performance of our saliency model is further improved by using a set of linear normalization-based probability distribution distance metrics as loss functions. As such, we can formulate saliency prediction as a probability distribution prediction task for global saliency inference instead of a typical pixel-wise regression problem. Experimental results on several challenging saliency benchmark datasets demonstrate that our DINet with proposed loss functions can achieve state-of-the-art performance with shorter inference time.
Predicting salient regions in natural images requires the detection of objects that are present in a scene. To develop robust representations for this challenging task, high-level visual features at multiple spatial scales must be extracted and augmented with contextual information. However, existing models aimed at explaining human fixation maps do not incorporate such a mechanism explicitly. Here we propose an approach based on a convolutional neural network pre-trained on a large-scale image classification task. The architecture forms an encoder-decoder structure and includes a module with multiple convolutional layers at different dilation rates to capture multi-scale features in parallel. Moreover, we combine the resulting representations with global scene information for accurately predicting visual saliency. Our model achieves competitive and consistent results across multiple evaluation metrics on two public saliency benchmarks and we demonstrate the effectiveness of the suggested approach on five datasets and selected examples. Compared to state of the art approaches, the network is based on a lightweight image classification backbone and hence presents a suitable choice for applications with limited computational resources, such as (virtual) robotic systems, to estimate human fixations across complex natural scenes.
In this work, we contribute to video saliency research in two ways. First, we introduce a new benchmark for predicting human eye movements during dynamic scene free-viewing, which is long-time urged in this field. Our dataset, named DHF1K (Dynamic Human Fixation), consists of 1K high-quality, elaborately selected video sequences spanning a large range of scenes, motions, object types and background complexity. Existing video saliency datasets lack variety and generality of common dynamic scenes and fall short in covering challenging situations in unconstrained environments. In contrast, DHF1K makes a significant leap in terms of scalability, diversity and difficulty, and is expected to boost video saliency modeling. Second, we propose a novel video saliency model that augments the CNN-LSTM network architecture with an attention mechanism to enable fast, end-to-end saliency learning. The attention mechanism explicitly encodes static saliency information, thus allowing LSTM to focus on learning more flexible temporal saliency representation across successive frames. Such a design fully leverages existing large-scale static fixation datasets, avoids overfitting, and significantly improves training efficiency and testing performance. We thoroughly examine the performance of our model, with respect to state-of-the-art saliency models, on three large-scale datasets (i.e., DHF1K, Hollywood2, UCF sports). Experimental results over more than 1.2K testing videos containing 400K frames demonstrate that our model outperforms other competitors.