Patch Is Not All You Need

Vision Transformers have achieved great success in computer visions, delivering exceptional performance across various tasks. However, their inherent reliance on sequential input enforces the manual partitioning of images into patch sequences, which disrupts the image's inherent structural and semantic continuity. To handle this, we propose a novel Pattern Transformer (Patternformer) to adaptively convert images to pattern sequences for Transformer input. Specifically, we employ the Convolutional Neural Network to extract various patterns from the input image, with each channel representing a unique pattern that is fed into the succeeding Transformer as a visual token. By enabling the network to optimize these patterns, each pattern concentrates on its local region of interest, thereby preserving its intrinsic structural and semantic information. Only employing the vanilla ResNet and Transformer, we have accomplished state-of-the-art performance on CIFAR-10 and CIFAR-100, and have achieved competitive results on ImageNet.

DistilXLSR: A Light Weight Cross-Lingual Speech Representation Model

Multilingual self-supervised speech representation models have greatly enhanced the speech recognition performance for low-resource languages, and the compression of these huge models has also become a crucial prerequisite for their industrial application. In this paper, we propose DistilXLSR, a distilled cross-lingual speech representation model. By randomly shuffling the phonemes of existing speech, we reduce the linguistic information and distill cross-lingual models using only English data. We also design a layer-jumping initialization method to fully leverage the teacher's pre-trained weights. Experiments on 2 kinds of teacher models and 15 low-resource languages show that our method can reduce the parameters by 50% while maintaining cross-lingual representation ability. Our method is proven to be generalizable to various languages/teacher models and has the potential to improve the cross-lingual performance of the English pre-trained models.

Inferring and Leveraging Parts from Object Shape for Improving Semantic Image Synthesis

Despite the progress in semantic image synthesis, it remains a challenging problem to generate photo-realistic parts from input semantic map. Integrating part segmentation map can undoubtedly benefit image synthesis, but is bothersome and inconvenient to be provided by users. To improve part synthesis, this paper presents to infer Parts from Object ShapE (iPOSE) and leverage it for improving semantic image synthesis. However, albeit several part segmentation datasets are available, part annotations are still not provided for many object categories in semantic image synthesis. To circumvent it, we resort to few-shot regime to learn a PartNet for predicting the object part map with the guidance of pre-defined support part maps. PartNet can be readily generalized to handle a new object category when a small number (e.g., 3) of support part maps for this category are provided. Furthermore, part semantic modulation is presented to incorporate both inferred part map and semantic map for image synthesis. Experiments show that our iPOSE not only generates objects with rich part details, but also enables to control the image synthesis flexibly. And our iPOSE performs favorably against the state-of-the-art methods in terms of quantitative and qualitative evaluation. Our code will be publicly available at https://github.com/csyxwei/iPOSE.

TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition

Text irregularities pose significant challenges to scene text recognizers. Thin-Plate Spline (TPS)-based rectification is widely regarded as an effective means to deal with them. Currently, the calculation of TPS transformation parameters purely depends on the quality of regressed text borders. It ignores the text content and often leads to unsatisfactory rectified results for severely distorted text. In this work, we introduce TPS++, an attention-enhanced TPS transformation that incorporates the attention mechanism to text rectification for the first time. TPS++ formulates the parameter calculation as a joint process of foreground control point regression and content-based attention score estimation, which is computed by a dedicated designed gated-attention block. TPS++ builds a more flexible content-aware rectifier, generating a natural text correction that is easier to read by the subsequent recognizer. Moreover, TPS++ shares the feature backbone with the recognizer in part and implements the rectification at feature-level rather than image-level, incurring only a small overhead in terms of parameters and inference time. Experiments on public benchmarks show that TPS++ consistently improves the recognition and achieves state-of-the-art accuracy. Meanwhile, it generalizes well on different backbones and recognizers. Code is at https://github.com/simplify23/TPS_PP.

CCLAP: Controllable Chinese Landscape Painting Generation via Latent Diffusion Model

With the development of deep generative models, recent years have seen great success of Chinese landscape painting generation. However, few works focus on controllable Chinese landscape painting generation due to the lack of data and limited modeling capabilities. In this work, we propose a controllable Chinese landscape painting generation method named CCLAP, which can generate painting with specific content and style based on Latent Diffusion Model. Specifically, it consists of two cascaded modules, i.e., content generator and style aggregator. The content generator module guarantees the content of generated paintings specific to the input text. While the style aggregator module is to generate paintings of a style corresponding to a reference image. Moreover, a new dataset of Chinese landscape paintings named CLAP is collected for comprehensive evaluation. Both the qualitative and quantitative results demonstrate that our method achieves state-of-the-art performance, especially in artfully-composed and artistic conception. Codes are available at https://github.com/Robin-WZQ/CCLAP.

ELITE: Encoding Visual Concepts into Textual Embeddings for Customized Text-to-Image Generation

Despite unprecedented ability in imaginary creation, large text-to-image models are further expected to express customized concepts. Existing works generally learn such concepts in an optimization-based manner, yet bringing excessive computation or memory burden. In this paper, we instead propose a learning-based encoder for fast and accurate concept customization, which consists of global and local mapping networks. In specific, the global mapping network separately projects the hierarchical features of a given image into multiple ``new'' words in the textual word embedding space, i.e., one primary word for well-editable concept and other auxiliary words to exclude irrelevant disturbances (e.g., background). In the meantime, a local mapping network injects the encoded patch features into cross attention layers to provide omitted details, without sacrificing the editability of primary concepts. We compare our method with prior optimization-based approaches on a variety of user-defined concepts, and demonstrate that our method enables more high-fidelity inversion and robust editability with a significantly faster encoding process. Our code will be publicly available at https://github.com/csyxwei/ELITE.

Position-Aware Contrastive Alignment for Referring Image Segmentation

Referring image segmentation aims to segment the target object described by a given natural language expression. Typically, referring expressions contain complex relationships between the target and its surrounding objects. The main challenge of this task is to understand the visual and linguistic content simultaneously and to find the referred object accurately among all instances in the image. Currently, the most effective way to solve the above problem is to obtain aligned multi-modal features by computing the correlation between visual and linguistic feature modalities under the supervision of the ground-truth mask. However, existing paradigms have difficulty in thoroughly understanding visual and linguistic content due to the inability to perceive information directly about surrounding objects that refer to the target. This prevents them from learning aligned multi-modal features, which leads to inaccurate segmentation. To address this issue, we present a position-aware contrastive alignment network (PCAN) to enhance the alignment of multi-modal features by guiding the interaction between vision and language through prior position information. Our PCAN consists of two modules: 1) Position Aware Module (PAM), which provides position information of all objects related to natural language descriptions, and 2) Contrastive Language Understanding Module (CLUM), which enhances multi-modal alignment by comparing the features of the referred object with those of related objects. Extensive experiments on three benchmarks demonstrate our PCAN performs favorably against the state-of-the-art methods. Our code will be made publicly available.

1st Place Solution for YouTubeVOS Challenge 2022: Referring Video Object Segmentation

The task of referring video object segmentation aims to segment the object in the frames of a given video to which the referring expressions refer. Previous methods adopt multi-stage approach and design complex pipelines to obtain promising results. Recently, the end-to-end method based on Transformer has proved its superiority. In this work, we draw on the advantages of the above methods to provide a simple and effective pipeline for RVOS. Firstly, We improve the state-of-the-art one-stage method ReferFormer to obtain mask sequences that are strongly correlated with language descriptions. Secondly, based on a reliable and high-quality keyframe, we leverage the superior performance of video object segmentation model to further enhance the quality and temporal consistency of the mask results. Our single model reaches 70.3 J &F on the Referring Youtube-VOS validation set and 63.0 on the test set. After ensemble, we achieve 64.1 on the final leaderboard, ranking 1st place on CVPR2022 Referring Youtube-VOS challenge. Code will be available at https://github.com/Zhiweihhh/cvpr2022-rvos-challenge.git.

Texts as Images in Prompt Tuning for Multi-Label Image Recognition

Prompt tuning has been employed as an efficient way to adapt large vision-language pre-trained models (e.g. CLIP) to various downstream tasks in data-limited or label-limited settings. Nonetheless, visual data (e.g., images) is by default prerequisite for learning prompts in existing methods. In this work, we advocate that the effectiveness of image-text contrastive learning in aligning the two modalities (for training CLIP) further makes it feasible to treat texts as images for prompt tuning and introduce TaI prompting. In contrast to the visual data, text descriptions are easy to collect, and their class labels can be directly derived. Particularly, we apply TaI prompting to multi-label image recognition, where sentences in the wild serve as alternatives to images for prompt tuning. Moreover, with TaI, double-grained prompt tuning (TaI-DPT) is further presented to extract both coarse-grained and fine-grained embeddings for enhancing the multi-label recognition performance. Experimental results show that our proposed TaI-DPT outperforms zero-shot CLIP by a large margin on multiple benchmarks, e.g., MS-COCO, VOC2007, and NUS-WIDE, while it can be combined with existing methods of prompting from images to improve recognition performance further. Code is released at https://github.com/guozix/TaI-DPT.

DSPGAN: a GAN-based universal vocoder for high-fidelity TTS by time-frequency domain supervision from DSP

Recent development of neural vocoders based on the generative adversarial neural network (GAN) has shown their advantages of generating raw waveform conditioned on mel-spectrogram with fast inference speed and lightweight networks. Whereas, it is still challenging to train a universal neural vocoder that can synthesize high-fidelity speech from various scenarios with unseen speakers, languages, and speaking styles. In this paper, we propose DSPGAN, a GAN-based universal vocoder for high-fidelity speech synthesis by applying the time-frequency domain supervision from digital signal processing (DSP). To eliminate the mismatch problem caused by the ground-truth spectrograms in training phase and the predicted spectrograms in inference phase, we leverage the mel-spectrogram extracted from the waveform generated by a DSP module, rather than the predicted mel-spectrogram from the Text-to-Speech (TTS) acoustic model, as the time-frequency domain supervision to the GAN-based vocoder. We also utilize sine excitation as the time-domain supervision to improve the harmonic modeling and eliminate various artifacts of the GAN-based vocoder. Experimental results show that DSPGAN significantly outperforms the compared approaches and can generate high-fidelity speech based on diverse data in TTS.