Multi-exit architectures consist of a backbone and branch classifiers that offer shortened inference pathways to reduce the run-time of deep neural networks. In this paper, we analyze different branching patterns that vary in their allocation of computational complexity for the branch classifiers. Constant-complexity branching keeps all branches the same, while complexity-increasing and complexity-decreasing branching place more complex branches later or earlier in the backbone respectively. Through extensive experimentation on multiple backbones and datasets, we find that complexity-decreasing branches are more effective than constant-complexity or complexity-increasing branches, which achieve the best accuracy-cost trade-off. We investigate a cause by using knowledge consistency to probe the effect of adding branches onto a backbone. Our findings show that complexity-decreasing branching yields the least disruption to the feature abstraction hierarchy of the backbone, which explains the effectiveness of the branching patterns.
Most of the existing work in one-stage referring expression comprehension (REC) mainly focuses on multi-modal fusion and reasoning, while the influence of other factors in this task lacks in-depth exploration. To fill this gap, we conduct an empirical study in this paper. Concretely, we first build a very simple REC network called SimREC, and ablate 42 candidate designs/settings, which covers the entire process of one-stage REC from network design to model training. Afterwards, we conduct over 100 experimental trials on three benchmark datasets of REC. The extensive experimental results not only show the key factors that affect REC performance in addition to multi-modal fusion, e.g., multi-scale features and data augmentation, but also yield some findings that run counter to conventional understanding. For example, as a vision and language (V&L) task, REC does is less impacted by language prior. In addition, with a proper combination of these findings, we can improve the performance of SimREC by a large margin, e.g., +27.12% on RefCOCO+, which outperforms all existing REC methods. But the most encouraging finding is that with much less training overhead and parameters, SimREC can still achieve better performance than a set of large-scale pre-trained models, e.g., UNITER and VILLA, portraying the special role of REC in existing V&L research.
Despite the exciting performance, Transformer is criticized for its excessive parameters and computation cost. However, compressing Transformer remains as an open problem due to its internal complexity of the layer designs, i.e., Multi-Head Attention (MHA) and Feed-Forward Network (FFN). To address this issue, we introduce Group-wise Transformation towards a universal yet lightweight Transformer for vision-and-language tasks, termed as LW-Transformer. LW-Transformer applies Group-wise Transformation to reduce both the parameters and computations of Transformer, while also preserving its two main properties, i.e., the efficient attention modeling on diverse subspaces of MHA, and the expanding-scaling feature transformation of FFN. We apply LW-Transformer to a set of Transformer-based networks, and quantitatively measure them on three vision-and-language tasks and six benchmark datasets. Experimental results show that while saving a large number of parameters and computations, LW-Transformer achieves very competitive performance against the original Transformer networks for vision-and-language tasks. To examine the generalization ability, we also apply our optimization strategy to a recently proposed image Transformer called Swin-Transformer for image classification, where the effectiveness can be also confirmed
Pixel synthesis is a promising research paradigm for image generation, which can well exploit pixel-wise prior knowledge for generation. However, existing methods still suffer from excessive memory footprint and computation overhead. In this paper, we propose a progressive pixel synthesis network towards efficient image generation, coined as PixelFolder. Specifically, PixelFolder formulates image generation as a progressive pixel regression problem and synthesizes images by a multi-stage paradigm, which can greatly reduce the overhead caused by large tensor transformations. In addition, we introduce novel pixel folding operations to further improve model efficiency while maintaining pixel-wise prior knowledge for end-to-end regression. With these innovative designs, we greatly reduce the expenditure of pixel synthesis, e.g., reducing 90% computation and 57% parameters compared to the latest pixel synthesis method called CIPS. To validate our approach, we conduct extensive experiments on two benchmark datasets, namely FFHQ and LSUN Church. The experimental results show that with much less expenditure, PixelFolder obtains new state-of-the-art (SOTA) performance on two benchmark datasets, i.e., 3.77 FID and 2.45 FID on FFHQ and LSUN Church, respectively. Meanwhile, PixelFolder is also more efficient than the SOTA methods like StyleGAN2, reducing about 74% computation and 36% parameters, respectively. These results greatly validate the effectiveness of the proposed PixelFolder.
Most existing Human-Object Interaction~(HOI) Detection methods rely heavily on full annotations with predefined HOI categories, which is limited in diversity and costly to scale further. We aim at advancing zero-shot HOI detection to detect both seen and unseen HOIs simultaneously. The fundamental challenges are to discover potential human-object pairs and identify novel HOI categories. To overcome the above challenges, we propose a novel end-to-end zero-shot HOI Detection (EoID) framework via vision-language knowledge distillation. We first design an Interactive Score module combined with a Two-stage Bipartite Matching algorithm to achieve interaction distinguishment for human-object pairs in an action-agnostic manner. Then we transfer the distribution of action probability from the pretrained vision-language teacher as well as the seen ground truth to the HOI model to attain zero-shot HOI classification. Extensive experiments on HICO-Det dataset demonstrate that our model discovers potential interactive pairs and enables the recognition of unseen HOIs. Finally, our method outperforms the previous SOTA by 8.92% on unseen mAP and 10.18% on overall mAP under UA setting, by 6.02% on unseen mAP and 9.1% on overall mAP under UC setting. Moreover, our method is generalizable to large-scale object detection data to further scale up the action sets. The source code will be available at: https://github.com/mrwu-mac/EoID.
In this paper, we propose a simple yet universal network termed SeqTR for visual grounding tasks, e.g., phrase localization, referring expression comprehension (REC) and segmentation (RES). The canonical paradigms for visual grounding often require substantial expertise in designing network architectures and loss functions, making them hard to generalize across tasks. To simplify and unify the modeling, we cast visual grounding as a point prediction problem conditioned on image and text inputs, where either the bounding box or binary mask is represented as a sequence of discrete coordinate tokens. Under this paradigm, visual grounding tasks are unified in our SeqTR network without task-specific branches or heads, e.g., the convolutional mask decoder for RES, which greatly reduces the complexity of multi-task modeling. In addition, SeqTR also shares the same optimization objective for all tasks with a simple cross-entropy loss, further reducing the complexity of deploying hand-crafted loss functions. Experiments on five benchmark datasets demonstrate that the proposed SeqTR outperforms (or is on par with) the existing state-of-the-arts, proving that a simple yet universal approach for visual grounding is indeed feasible.
Recently, Vision Transformer (ViT) has achieved remarkable success in several computer vision tasks. The progresses are highly relevant to the architecture design, then it is worthwhile to propose Transformer Architecture Search (TAS) to search for better ViTs automatically. However, current TAS methods are time-consuming and existing zero-cost proxies in CNN do not generalize well to the ViT search space according to our experimental observations. In this paper, for the first time, we investigate how to conduct TAS in a training-free manner and devise an effective training-free TAS (TF-TAS) scheme. Firstly, we observe that the properties of multi-head self-attention (MSA) and multi-layer perceptron (MLP) in ViTs are quite different and that the synaptic diversity of MSA affects the performance notably. Secondly, based on the observation, we devise a modular strategy in TF-TAS that evaluates and ranks ViT architectures from two theoretical perspectives: synaptic diversity and synaptic saliency, termed as DSS-indicator. With DSS-indicator, evaluation results are strongly correlated with the test accuracies of ViT models. Experimental results demonstrate that our TF-TAS achieves a competitive performance against the state-of-the-art manually or automatically design ViT architectures, and it promotes the searching efficiency in ViT search space greatly: from about $24$ GPU days to less than $0.5$ GPU days. Moreover, the proposed DSS-indicator outperforms the existing cutting-edge zero-cost approaches (e.g., TE-score and NASWOT).
This paper proposes an Any-time super-Resolution Method (ARM) to tackle the over-parameterized single image super-resolution (SISR) models. Our ARM is motivated by three observations: (1) The performance of different image patches varies with SISR networks of different sizes. (2) There is a tradeoff between computation overhead and performance of the reconstructed image. (3) Given an input image, its edge information can be an effective option to estimate its PSNR. Subsequently, we train an ARM supernet containing SISR subnets of different sizes to deal with image patches of various complexity. To that effect, we construct an Edge-to-PSNR lookup table that maps the edge score of an image patch to the PSNR performance for each subnet, together with a set of computation costs for the subnets. In the inference, the image patches are individually distributed to different subnets for a better computation-performance tradeoff. Moreover, each SISR subnet shares weights of the ARM supernet, thus no extra parameters are introduced. The setting of multiple subnets can well adapt the computational cost of SISR model to the dynamically available hardware resources, allowing the SISR task to be in service at any time. Extensive experiments on resolution datasets of different sizes with popular SISR networks as backbones verify the effectiveness and the versatility of our ARM. The source code is available at \url{https://github.com/chenbong/ARM-Net}.
Recently, automatic video captioning has attracted increasing attention, where the core challenge lies in capturing the key semantic items, like objects and actions as well as their spatial-temporal correlations from the redundant frames and semantic content. To this end, existing works select either the key video clips in a global level~(across multi frames), or key regions within each frame, which, however, neglect the hierarchical order, i.e., key frames first and key regions latter. In this paper, we propose a novel joint-hierarchical attention model for video captioning, which embeds the key clips, the key frames and the key regions jointly into the captioning model in a hierarchical manner. Such a joint-hierarchical attention model first conducts a global selection to identify key frames, followed by a Gumbel sampling operation to identify further key regions based on the key frames, achieving an accurate global-to-local feature representation to guide the captioning. Extensive quantitative evaluations on two public benchmark datasets MSVD and MSR-VTT demonstrates the superiority of the proposed method over the state-of-the-art methods.
Echocardiography is widely used to clinical practice for diagnosis and treatment, e.g., on the common congenital heart defects. The traditional manual manipulation is error-prone due to the staff shortage, excess workload, and less experience, leading to the urgent requirement of an automated computer-aided reporting system to lighten the workload of ultrasonologists considerably and assist them in decision making. Despite some recent successful attempts in automatical medical report generation, they are trapped in the ultrasound report generation, which involves unstructured-view images and topic-related descriptions. To this end, we investigate the task of the unstructured-view topic-related ultrasound report generation, and propose a novel factored attention and embedding model (termed FAE-Gen). The proposed FAE-Gen mainly consists of two modules, i.e., view-guided factored attention and topic-oriented factored embedding, which 1) capture the homogeneous and heterogeneous morphological characteristic across different views, and 2) generate the descriptions with different syntactic patterns and different emphatic contents for different topics. Experimental evaluations are conducted on a to-be-released large-scale clinical cardiovascular ultrasound dataset (CardUltData). Both quantitative comparisons and qualitative analysis demonstrate the effectiveness and the superiority of FAE-Gen over seven commonly-used metrics.