Attentional Feature Fusion

Dai, Yawen; Gieseke, Fabian; Oehmcke, Stefan; Wu, Yiquan; Barnard, Kobus

doi:10.1109/wacv48630.2021.00360

Cited by 404 publications

(145 citation statements)

References 30 publications

Supporting

Mentioning

142

Contrasting

Order By: Relevance

“…However, we simply concatenate the multilevel feature maps in the RF module as a feature fusion procedure, where the relationship among these maps is seldom considered. As we know, various fusion approaches, aiming to flexibly exploit the feature information, have successfully been presented in different scenarios [46–48]. Moreover, an attention mechanism has also been proved as a useful fusion strategy in image denoising as mentioned before.…”

Section: Discussionmentioning

confidence: 99%

Multiscale residual fusion network for image denoising

Yao

Tang

Sun

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

Deep‐learning methods have been developed in recent years and have achieved dramatic improvements for image denoising. The existing deep‐learning methods can be conducted using two major models: Encoder–decoder and high‐resolution, where the high‐resolution model has superior resolution ability for detail description and restoration. In this study, a high‐resolution‐based network called multiscale residual fusion network (MRF‐Net) is proposed, which employed the spatial and contextual information of images. In detail, dilated convolution layers are used to enlarge the network's receptive field and learned sufficient features in a multiscale feature extracting module. The function of dilated convolution is reinterpreted here and it is viewed as a complex downsampling operation. Therefore, multiscale feature analysis could be performed in the proposed network by dilated convolution. Multilevel feature maps are sequentially obtained through a residual projection module, where considerable contextual and spatial information was collected from the multiscale features. In a residual fusion module, all maps were aggregated to generate a residual image effectively for noise removal. Experiments demonstrated that the MRF‐Net outperformed several state‐of‐the‐art model‐based and deep‐learning methods in both blind and non‐blind image denoising tests. Meanwhile, ablation studies were executed to verify the denoising performance of each module. Moreover, this method exhibited high computational efficiency, thus demonstrating its practicability.

show abstract

Section: Discussionmentioning

confidence: 99%

Multiscale residual fusion network for image denoising

Yao

Tang

Sun

et al. 2021

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…Finally, we fuse the resized features by Attentional Feature Fusion (AFF [8]): Ψ im vt = AFF(Ψ im vol , Ψ im tex ). Ψ im vt has the same size as Ψ im vol .…”

Section: Hybrid Volumetric-textural Renderingmentioning

confidence: 99%

“…We compare two methods to fuse the volumetric and textural features as discussed at Sec. 3.4 by concatenation (Concat) and Attentional Feature Fusion (AFF [8]) on two datasets, R1 and R2 (about 12,000 frames in training, 3,000 frames in testing). We test the performances on novel poses.…”

Section: B2 Ablation Studymentioning

confidence: 99%

“…In addition, we extract 2D textural features from our latents on the UV manifold for realistic image synthesis following the spirit of Deferred Neural Rendering (DNR) [66,54,17]. We fuse the 3D and 2D features by utilizing Attentional Feature Fusion (AFF [8]), and finally use a 2D GAN-based [9] textural rendering network (TexRenderer) to decode and supersample them into realistic images. Though TexRenderer works in image space, it is able to incorporate the rasterized volumetric features for geometry-aware rendering.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

HVTR: Hybrid Volumetric-Textural Rendering for Human Avatars

Hu¹,

Chen²,

Zheng³

et al. 2021

Preprint

View full text Add to dashboard Cite

We propose a novel neural rendering pipeline, Hybrid Volumetric-Textural Rendering (HVTR), which synthesizes virtual human avatars from arbitrary poses efficiently and at high quality. First, we learn to encode articulated human motions on a dense UV manifold of the human body surface. To handle complicated motions (e.g., self-occlusions), we then leverage the encoded information on the UV manifold to construct a 3D volumetric representation based on a dynamic pose-conditioned neural radiance field. While this allows us to represent 3D geometry with changing topology, volumetric rendering is computationally heavy. Hence we employ only a rough volumetric representation using a pose-conditioned downsampled neural radiance field (PD-NeRF), which we can render efficiently at low resolutions. In addition, we learn 2D textural features that are fused with rendered volumetric features in image space. The key advantage of our approach is that we can then convert the fused features into a high resolution, high quality avatar by a fast GAN-based textural renderer. We demonstrate that hybrid rendering enables HVTR to handle complicated motions, render high quality avatars under usercontrolled poses/shapes and even loose clothing, and most importantly, be fast at inference time. Our experimental results also demonstrate state-of-the-art quantitative results. More results are available at our project page: https: // www. cs. umd. edu/ ~taohu/ hvtr/ .

show abstract

“…Another contemporary approach includes fusing channel and spatial attention (convolutional block attention module or CBAM) [21]. The literature [13] modified the feature fusion of YOLOv5 to AAF (Attention Feature Fusion) [22].…”

Section: Introductionmentioning

confidence: 99%

Attention-Enhanced One-Stage Algorithm for Traffic Sign Detection and Recognition

et al. 2022

View full text Add to dashboard Cite

Detection and recognition of traffic signs are the keys to creating advanced driving assistance systems. Making highly precise maps also requires the identification and extraction of such road elements as traffic signs. Traditional detection and recognition methods no longer meet today’s needs, and object recognition algorithms based on deep learning have become the mainstream solution. However, current algorithms have limitations. The recognition speed of one-stage strategy algorithms is fast, but recognition accuracy is not satisfactory especially for small objects. The accuracy of two-stage algorithms is higher, but the recognition speed is extremely slow. This paper solves these problems with a proposed parallel attention convolution module, a channel attention pyramid network, and a loss function diagonal and center point IoU based on the YOLOv3 algorithm. The improved models in this paper are compared with SSD, YOLOv3, and Faster RCNN. Experimental results show that the proposed models have some improvement over the above models: the mAP of the models with PACM, CAFPN, and DCPIoU was 76.02%, compared with SSD300, SSD500, Faster RCNN, and YOLOv3, which had improvements of 9.27%, 6.93%, 2.94, and 5.3%, respectively. And the FPS of the improved model is basically the same as the original YOLOv3, without reducing the real-time performance.

show abstract

Attentional Feature Fusion

Cited by 404 publications

References 30 publications

Multiscale residual fusion network for image denoising

Multiscale residual fusion network for image denoising

HVTR: Hybrid Volumetric-Textural Rendering for Human Avatars

Attention-Enhanced One-Stage Algorithm for Traffic Sign Detection and Recognition

Contact Info

Product

Resources

About