Video Paragraph Captioning aims to generate a multi-sentence description of an untrimmed video with multiple temporal event locations in a coherent storytelling. Following the human perception process, where the scene is effectively understood by decomposing it into visual (e.g. human, animal) and non- visual components (e.g. action, relations) under the mutual influence of vision and language, we first propose a visual- linguistic (VL) feature. In the proposed VL feature, the scene is modeled by three modalities including (i) a global visual environment; (ii) local visual main agents; (iii) linguistic scene elements. We then introduce an autoregressive Transformer- in-Transformer (TinT) to simultaneously capture the semantic coherence of intra- and inter-event contents within a video. Finally, we present a new VL contrastive loss function to guarantee the learnt embedding features are consistent with the captions semantics. Comprehensive experiments and extensive ablation studies on the ActivityNet Captions and YouCookII datasets show that the proposed Visual-Linguistic Transformer- in-Transform (VLTinT) outperforms previous state-of-the-art methods in terms of accuracy and diversity. The source code is made publicly available at: https://github.com/UARK-AICV/ VLTinT.
more »
« less
Vidtr: Video transformer without convolutions
We introduce Video Transformer (VidTr) with separableattention for video classification. Comparing with commonly
used 3D networks, VidTr is able to aggregate spatiotemporal information via stacked attentions and provide better performance with higher efficiency. We first introduce the vanilla video transformer and show that transformer
module is able to perform spatio-temporal modeling from raw pixels, but with heavy memory usage. We then present VidTr which reduces the memory cost by 3.3× while keeping the same performance. To further optimize the model,
we propose the standard deviation based topK pooling for attention (pooltopK std), which reduces the computation by dropping non-informative features along temporal dimension. VidTr achieves state-of-the-art performance on five commonly used datasets with lower computational requirement, showing both the efficiency and effectiveness of our design. Finally, error analysis and visualization show that VidTr is especially good at predicting actions that require long-term temporal reasoning.
more »
« less
- Award ID(s):
- 1763827
- PAR ID:
- 10329452
- Date Published:
- Journal Name:
- Proceedings of the IEEE International Conference on Computer Vision
- ISSN:
- 1550-5499
- Page Range / eLocation ID:
- 13577 - 13587
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Video Paragraph Captioning aims to generate a multi-sentence description of an untrimmed video with multiple temporal event locations in a coherent storytelling. Following the human perception process, where the scene is effectively understood by decomposing it into visual (e.g. human, animal) and non-visual components (e.g. action, relations) under the mutual influence of vision and language, we first propose a visual-linguistic (VL) feature. In the proposed VL feature, the scene is modeled by three modalities including (i) a global visual environment; (ii) local visual main agents; (iii) linguistic scene elements. We then introduce an autoregressive Transformer-in-Transformer (TinT) to simultaneously capture the semantic coherence of intra- and inter-event contents within a video. Finally, we present a new VL contrastive loss function to guarantee the learnt embedding features are consistent with the captions semantics. Comprehensive experiments and extensive ablation studies on the ActivityNet Captions and YouCookII datasets show that the proposed Visual-Linguistic Transformer-in-Transform (VLTinT) outperforms previous state-of-the-art methods in terms of accuracy and diversity. The source code is made publicly available at: https://github.com/UARK-AICV/VLTinT.more » « less
-
Stochastic computing (SC) reduces the complexity of computation by representing numbers with long streams of independent bits. However, increasing performance in SC comes with either an increase in area or a loss in accuracy. Processing in memory (PIM) computes data in-place while having high memory density and supporting bit-parallel operations with low energy consumption. In this article, we propose COSMO, an architecture for co mputing with s tochastic numbers in me mo ry, which enables SC in memory. The proposed architecture is general and can be used for a wide range of applications. It is a highly dense and parallel architecture that supports most SC encodings and operations in memory. It maximizes the performance and energy efficiency of SC by introducing several innovations: (i) in-memory parallel stochastic number generation, (ii) efficient implication-based logic in memory, (iii) novel memory bit line segmenting, (iv) a new memory-compatible SC addition operation, and (v) enabling flexible block allocation. To show the generality and efficiency of our stochastic architecture, we implement image processing, deep neural networks (DNNs), and hyperdimensional (HD) computing on the proposed hardware. Our evaluations show that running DNN inference on COSMO is 141× faster and 80× more energy efficient as compared to GPU.more » « less
-
In this paper, we aim at synthesizing a free-viewpoint video of an arbitrary human performance using sparse multi-view cameras. Recently, several works have addressed this problem by learning person-specific neural radiance fields (NeRF) to capture the appearance of a particular human. In parallel, some work proposed to use pixel-aligned features to generalize radiance fields to arbitrary new scenes and objects. Adopting such generalization approaches to humans, however, is highly challenging due to the heavy occlusions and dynamic articulations of body parts. To tackle this, we propose Neural Human Performer, a novel approach that learns generalizable neural radiance fields based on a parametric human body model for robust performance capture. Specifically, we first introduce a temporal transformer that aggregates tracked visual features based on the skeletal body motion over time. Moreover, a multi-view transformer is proposed to perform cross-attention between the temporally-fused features and the pixel-aligned features at each time step to integrate observations on the fly from multiple views. Experiments on the ZJU-MoCap and AIST datasets show that our method significantly outperforms recent generalizable NeRF methods on unseen identities and poses. The video results and code are available at https://youngjoongunc.github.io/nhp.more » « less
-
null (Ed.)Attention mechanisms have led to many breakthroughs in sequential data modeling but have yet to be incorporated into any generative algorithms for molecular design. Here we explore the impact of adding self-attention layers to generative β -VAE models and show that those with attention are able to learn a complex “molecular grammar” while improving performance on downstream tasks such as accurately sampling from the latent space (“model memory”) or exploring novel chemistries not present in the training data. There is a notable relationship between a model's architecture, the structure of its latent memory and its performance during inference. We demonstrate that there is an unavoidable tradeoff between model exploration and validity that is a function of the complexity of the latent memory. However, novel sampling schemes may be used that optimize this tradeoff. We anticipate that attention will play an important role in future molecular design algorithms that can make efficient use of the detailed molecular substructures learned by the transformer.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/ICCV48922.2021.01332
