Existing face forgery detection methods achieve promising performance when training and testing forgery data are from identical manipulation types, while they fail to generalize well to unseen samples. In this paper, we experimentally investigate and find that the poor generalization of the methods mainly arises from their overfitting on the known fake patterns. Excessively focused on seen fakes, those detectors fail to effectively learn image-intrinsic information and the distributional disparity between real and fake images. Then, to address this issue, we redefine fake learning as real-fake distributional disparity learning. We propose a novel deepfake detection framework learning distributional disparity based on the differentiated reconstruction on real and fake images for improved generalization. Specifically, distributional disparity learning on differentiated reconstruction of the real and fake images, enforces the model to learn image-invariant intrinsic representations. The reconstruction on real and fake images forces the decoders to learn the distribution of real and fake images, respectively. Moreover, to avoid the influence from the specificalization of the known fake patterns, we further propose the information interaction learning on the encoded intrinsic information and the pixel disparity between the input image and its reconstruction to distinguish face forgeries that are even unknown. Extensive experiments on large-scale benchmark datasets demonstrated the effectiveness of addressing the overfitting issue of the classification network, and verified the superior performance of our method.

Recent mainstream image captioning methods usually adopt two-stage captioners, i.e., calculating the object features of the given image by a pre-trained detector and then feeding them into a language model to generate the descriptive sentences. However, such a two-stage procedure will lead to a task-based information gap that decreases the performance of the captioners, because the object features learned from the detection task are suboptimal representations and cannot provide all the necessary information for subsequent sentence generation. Besides, the object features are usually represented by the last pooling features of the detector that lose the local details of images. In this paper, we propose a novel One-Stage Image Captioner using dynamic multi-sight embedding and alignment, called SeaCap, which directly transforms input images into descriptive sentences in one stage to eliminate the information gap. Specifically, to obtain rich features, we use the Swin Transformer to capture multi-level features, followed by a sights alignment module to alleviate the vision confusion, and then feed them into a novel dynamic multi-sight embedding module to exploit both the global structure and local texture of input images. To enhance the global modeling capacity of the visual encoder, we propose a new dual-dimensional refining module to non-locally model the interaction of the embedded features. As a result, SeaCap can obtain rich and useful information to improve the performance of the captioner. Extensive comparisons on the benchmark MS-COCO, Flickr8K and Flickr30 K datasets verified the superior performance of our method.

Transformer-based approaches to image captioning have shown great success by utilizing long-term dependency for visual embedding. However, their coarse long-term dependency, using the multi-head self-attention mechanism to capture the contextual interactions between the visual tokens on the time step and (or) embedded dimension, fail to distinguish fine-grained features of local partition. In this case, some similar features are captured, which leads to feature redundancy that decreases the performance. To respond to this issue, this paper proposes a novel image captioner embedding visual cross-partition dependency, dubbed CropCap. Specifically, the visual sequence generated from the Swin Transformer-based pre-embedding network is fed into the proposed cross-partition dependency module to refinedly model the interaction between partial representations on both the time step and embedded dimension. Furthermore, we formulaically reason the proposed cross-partition dependency, and theoretically prove its correctness. Extensive comparisons on the benchmark MS-COCO dataset demonstrated the effectiveness addressing the information redundancy issue, and verified the superior performance of our method.

Image captioning is still a challenging task aiming at describing the contents of image by words. Current image caption methods usually assume the object relation to be important if the semantic and spatial geometric relationships between objects are close and large, but the relations meeting this assumption are not necessarily important to describe the contents of image in a fine-grained way. That is, the importance of fine-grained object relations is not properly taken into account. Besides, current Transformer based image caption models also fail to consider the importance of fine-grained objects, since they generate all the words of a sentence at one time, which cannot Figure out which objects are more important and vice versa. In this paper, we propose a novel Fine-grained Adaptive Object Transformer (FineFormer) network, which can jointly discover the importance of fine-grained objects and object relations for image captioning. Specifically, a new concept of adaptive soft-foreground attention is proposed to highlight the fine-grained objects dominating the descriptive contents. To characterize and calculate the important relations between fine-grained objects, we also propose an adaptive object relation attention to refine the object relation from the generation process of relation. As such, FineFormer can describe the contents of image more accurately, by reducing the interference of unimportant objects in the background. Extensive experiments on the highly-competitive MS-COCO dataset demonstrated the superiority of our FineFormer.

Although deep learning has achieved great success in the field of medical image processing, the existing deep learning based medical image segmentation solutions still cannot obtain satisfactory performances for abdominal small organs and lesions due to their small object size and shape-variability. In this work, a Deeply Supervised Multi-Scale U-Net (DSMS U-Net) is proposed for more accurate segmentation performances on abdominal small organs images. DSMS U-Net integrate the existing U-Net model with a restoration decoder module and some multi-scale convolution modules. Our experiment results demonstrate that the proposed DSMS U-Net approach has much better segmentation performances than the state-of-the-art baselines.