Spherical World-Locking for Audio-Visual Localization in Egocentric Videos

Egocentric videos provide comprehensive contexts for user and scene understanding, spanning multisensory perception to behavioral interaction. We propose Spherical World-Locking (SWL) as a general framework for egocentric scene representation, which implicitly transforms multisensory streams with respect to measurements of head orientation. Compared to conventional head-locked egocentric representations with a 2D planar field-of-view, SWL effectively offsets challenges posed by self-motion, allowing for improved spatial synchronization between input modalities. Using a set of multisensory embeddings on a worldlocked sphere, we design a unified encoder-decoder transformer architecture that preserves the spherical structure of the scene representation, without requiring expensive projections between image and world coordinate systems. We evaluate the effectiveness of the proposed framework on multiple benchmark tasks for egocentric video understanding, including audio-visual active speaker localization, auditory spherical source localization, and behavior anticipation in everyday activities.

Dense 2D-3D Indoor Prediction with Sound via Aligned Cross-Modal Distillation

Sound can convey significant information for spatial reasoning in our daily lives. To endow deep networks with such ability, we address the challenge of dense indoor prediction with sound in both 2D and 3D via cross-modal knowledge distillation. In this work, we propose a Spatial Alignment via Matching (SAM) distillation framework that elicits local correspondence between the two modalities in vision-to-audio knowledge transfer. SAM integrates audio features with visually coherent learnable spatial embeddings to resolve inconsistencies in multiple layers of a student model. Our approach does not rely on a specific input representation, allowing for flexibility in the input shapes or dimensions without performance degradation. With a newly curated benchmark named Dense Auditory Prediction of Surroundings (DAPS), we are the first to tackle dense indoor prediction of omnidirectional surroundings in both 2D and 3D with audio observations. Specifically, for audio-based depth estimation, semantic segmentation, and challenging 3D scene reconstruction, the proposed distillation framework consistently achieves state-of-the-art performance across various metrics and backbone architectures.

Panoramic Vision Transformer for Saliency Detection in 360° Videos

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.

Multimodal Knowledge Alignment with Reinforcement Learning

Large language models readily adapt to novel settings, even without task-specific training data. Can their zero-shot capacity be extended to multimodal inputs? In this work, we propose ESPER which extends language-only zero-shot models to unseen multimodal tasks, like image and audio captioning. Our key novelty is to use reinforcement learning to align multimodal inputs to language model generations without direct supervision: for example, in the image case our reward optimization relies only on cosine similarity derived from CLIP, and thus requires no additional explicitly paired (image, caption) data. Because the parameters of the language model are left unchanged, the model maintains its capacity for zero-shot generalization. Experiments demonstrate that ESPER outperforms baselines and prior work on a variety of zero-shot tasks; these include a new benchmark we collect+release, ESP dataset, which tasks models with generating several diversely-styled captions for each image.

Pano-AVQA: Grounded Audio-Visual Question Answering on 360$^\circ$ Videos

360$^\circ$ videos convey holistic views for the surroundings of a scene. It provides audio-visual cues beyond pre-determined normal field of views and displays distinctive spatial relations on a sphere. However, previous benchmark tasks for panoramic videos are still limited to evaluate the semantic understanding of audio-visual relationships or spherical spatial property in surroundings. We propose a novel benchmark named Pano-AVQA as a large-scale grounded audio-visual question answering dataset on panoramic videos. Using 5.4K 360$^\circ$ video clips harvested online, we collect two types of novel question-answer pairs with bounding-box grounding: spherical spatial relation QAs and audio-visual relation QAs. We train several transformer-based models from Pano-AVQA, where the results suggest that our proposed spherical spatial embeddings and multimodal training objectives fairly contribute to a better semantic understanding of the panoramic surroundings on the dataset.

Video Summarization through Human Detection on a Social Robot

In this paper, we propose a novel video summarization system which captures images via a social robot's camera but processes images on a server. The system helps remote family members easily be aware of their seniors' daily activities via summaries. The system utilizes two vision-based algorithms, one for pose estimation and the other for human detection, to locate people in frames to guide the robot through people tracking and filter out improper frames including the ones without a person or blurred, or with a person but too small or not at the center of the frame. The system utilizes a video summarization method to select keyframes by balancing the representativeness and diversity. We conduct experiments of the system through three in-the-wild studies and evaluate the performance through human subject studies. Experimental results show that the users of the system think the system is promising and useful for their needs.