Yu Gao, Lixue Gong, Qiushan Guo, Xiaoxia Hou, Zhichao Lai, Fanshi Li, Liang Li, Xiaochen Lian, Chao Liao, Liyang Liu, Wei Liu, Yichun Shi, Shiqi Sun, Yu Tian, Zhi Tian, Peng Wang, Rui Wang, Xuanda Wang, Xun Wang, Ye Wang, Guofeng Wu, Jie Wu, Xin Xia, Xuefeng Xiao, Zhonghua Zhai, Xinyu Zhang, Qi Zhang, Yuwei Zhang, Shijia Zhao, Jianchao Yang, Weilin Huang
We present Seedream 3.0, a high-performance Chinese-English bilingual image generation foundation model. We develop several technical improvements to address existing challenges in Seedream 2.0, including alignment with complicated prompts, fine-grained typography generation, suboptimal visual aesthetics and fidelity, and limited image resolutions. Specifically, the advancements of Seedream 3.0 stem from improvements across the entire pipeline, from data construction to model deployment. At the data stratum, we double the dataset using a defect-aware training paradigm and a dual-axis collaborative data-sampling framework. Furthermore, we adopt several effective techniques such as mixed-resolution training, cross-modality RoPE, representation alignment loss, and resolution-aware timestep sampling in the pre-training phase. During the post-training stage, we utilize diversified aesthetic captions in SFT, and a VLM-based reward model with scaling, thereby achieving outputs that well align with human preferences. Furthermore, Seedream 3.0 pioneers a novel acceleration paradigm. By employing consistent noise expectation and importance-aware timestep sampling, we achieve a 4 to 8 times speedup while maintaining image quality. Seedream 3.0 demonstrates significant improvements over Seedream 2.0: it enhances overall capabilities, in particular for text-rendering in complicated Chinese characters which is important to professional typography generation. In addition, it provides native high-resolution output (up to 2K), allowing it to generate images with high visual quality.
Team Seedream, :, Yunpeng Chen, Yu Gao, Lixue Gong, Meng Guo, Qiushan Guo, Zhiyao Guo, Xiaoxia Hou, Weilin Huang, Yixuan Huang, Xiaowen Jian, Huafeng Kuang, Zhichao Lai, Fanshi Li, Liang Li, Xiaochen Lian, Chao Liao, Liyang Liu, Wei Liu, Yanzuo Lu, Zhengxiong Luo, Tongtong Ou, Guang Shi, Yichun Shi, Shiqi Sun, Yu Tian, Zhi Tian, Peng Wang, Rui Wang, Xun Wang, Ye Wang, Guofeng Wu, Jie Wu, Wenxu Wu, Yonghui Wu, Xin Xia, Xuefeng Xiao, Shuang Xu, Xin Yan, Ceyuan Yang, Jianchao Yang, Zhonghua Zhai, Chenlin Zhang, Heng Zhang, Qi Zhang, Xinyu Zhang, Yuwei Zhang, Shijia Zhao, Wenliang Zhao, Wenjia Zhu
We introduce Seedream 4.0, an efficient and high-performance multimodal image generation system that unifies text-to-image (T2I) synthesis, image editing, and multi-image composition within a single framework. We develop a highly efficient diffusion transformer with a powerful VAE which also can reduce the number of image tokens considerably. This allows for efficient training of our model, and enables it to fast generate native high-resolution images (e.g., 1K-4K). Seedream 4.0 is pretrained on billions of text-image pairs spanning diverse taxonomies and knowledge-centric concepts. Comprehensive data collection across hundreds of vertical scenarios, coupled with optimized strategies, ensures stable and large-scale training, with strong generalization. By incorporating a carefully fine-tuned VLM model, we perform multi-modal post-training for training both T2I and image editing tasks jointly. For inference acceleration, we integrate adversarial distillation, distribution matching, and quantization, as well as speculative decoding. It achieves an inference time of up to 1.8 seconds for generating a 2K image (without a LLM/VLM as PE model). Comprehensive evaluations reveal that Seedream 4.0 can achieve state-of-the-art results on both T2I and multimodal image editing. In particular, it demonstrates exceptional multimodal capabilities in complex tasks, including precise image editing and in-context reasoning, and also allows for multi-image reference, and can generate multiple output images. This extends traditional T2I systems into an more interactive and multidimensional creative tool, pushing the boundary of generative AI for both creativity and professional applications. We further scale our model and data as Seedream 4.5. Seedream 4.0 and Seedream 4.5 are accessible on Volcano Engine https://www.volcengine.com/experience/ark?launch=seedream.
Jie Wu, Yu Gao, Zilyu Ye, Ming Li, Liang Li, Hanzhong Guo, Jie Liu, Zeyue Xue, Xiaoxia Hou, Wei Liu, Yan Zeng, Weilin Huang
Reward Models (RMs) are critical for improving generation models via Reinforcement Learning (RL), yet the RM scaling paradigm in visual generation remains largely unexplored. It primarily due to fundamental limitations in existing approaches: CLIP-based RMs suffer from architectural and input modality constraints, while prevalent Bradley-Terry losses are fundamentally misaligned with the next-token prediction mechanism of Vision-Language Models (VLMs), hindering effective scaling. More critically, the RLHF optimization process is plagued by Reward Hacking issue, where models exploit flaws in the reward signal without improving true quality. To address these challenges, we introduce RewardDance, a scalable reward modeling framework that overcomes these barriers through a novel generative reward paradigm. By reformulating the reward score as the model's probability of predicting a "yes" token, indicating that the generated image outperforms a reference image according to specific criteria, RewardDance intrinsically aligns reward objectives with VLM architectures. This alignment unlocks scaling across two dimensions: (1) Model Scaling: Systematic scaling of RMs up to 26 billion parameters; (2) Context Scaling: Integration of task-specific instructions, reference examples, and chain-of-thought (CoT) reasoning. Extensive experiments demonstrate that RewardDance significantly surpasses state-of-the-art methods in text-to-image, text-to-video, and image-to-video generation. Crucially, we resolve the persistent challenge of "reward hacking": Our large-scale RMs exhibit and maintain high reward variance during RL fine-tuning, proving their resistance to hacking and ability to produce diverse, high-quality outputs. It greatly relieves the mode collapse problem that plagues smaller models.
Ceyuan Yang, Zhijie Lin, Yang Zhao, Fei Xiao, Hao He, Qi Zhao, Chaorui Deng, Kunchang Li, Zihan Ding, Yuwei Guo, Fuyun Wang, Fangqi Zhu, Xiaonan Nie, Shenhan Zhu, Shanchuan Lin, Hongsheng Li, Weilin Huang, Guang Shi, Haoqi Fan
We present Omni, a unified multimodal model natively trained on diverse modalities, including text, images, videos, 3D geometry, and hidden representations. We find that such training enables Context Unrolling, where the model explicitly reasons across multiple modal representations before producing predictions. This process enables the model to aggregate complementary information across heterogeneous modalities, facilitating a more faithful approximation of the shared multimodal knowledge manifold and improving downstream reasoning fidelity. As a result, Omni achieves strong performance on both multimodal generation and understanding benchmarks, while demonstrating advanced multimodal reasoning capabilities, including in-context generation of text, image, video, and 3D geometry.
Xun Wang, Haozhi Zhang, Weilin Huang, Matthew R. Scott
Mining informative negative instances are of central importance to deep metric learning (DML), however this task is intrinsically limited by mini-batch training, where only a mini-batch of instances is accessible at each iteration. In this paper, we identify a "slow drift" phenomena by observing that the embedding features drift exceptionally slow even as the model parameters are updating throughout the training process. This suggests that the features of instances computed at preceding iterations can be used to considerably approximate their features extracted by the current model. We propose a cross-batch memory (XBM) mechanism that memorizes the embeddings of past iterations, allowing the model to collect sufficient hard negative pairs across multiple mini-batches - even over the whole dataset. Our XBM can be directly integrated into a general pair-based DML framework, where the XBM augmented DML can boost performance considerably. In particular, without bells and whistles, a simple contrastive loss with our XBM can have large R@1 improvements of 12%-22.5% on three large-scale image retrieval datasets, surpassing the most sophisticated state-of-the-art methods, by a large margin. Our XBM is conceptually simple, easy to implement - using several lines of codes, and is memory efficient - with a negligible 0.2 GB extra GPU memory. Code is available at: https://github.com/MalongTech/research-xbm.
Fengyuan Shi, Weilin Huang, Limin Wang
Video grounding aims to localize the corresponding video moment in an untrimmed video given a language query. Existing methods often address this task in an indirect way, by casting it as a proposal-and-match or fusion-and-detection problem. Solving these surrogate problems often requires sophisticated label assignment during training and hand-crafted removal of near-duplicate results. Meanwhile, existing works typically focus on sparse video grounding with a single sentence as input, which could result in ambiguous localization due to its unclear description. In this paper, we tackle a new problem of dense video grounding, by simultaneously localizing multiple moments with a paragraph as input. From a perspective on video grounding as language conditioned regression, we present an end-to-end parallel decoding paradigm by re-purposing a Transformer-alike architecture (PRVG). The key design in our PRVG is to use languages as queries, and directly regress the moment boundaries based on language-modulated visual representations. Thanks to its simplicity in design, our PRVG framework can be applied in different testing schemes (sparse or dense grounding) and allows for efficient inference without any post-processing technique. In addition, we devise a robust proposal-level attention loss to guide the training of PRVG, which is invariant to moment duration and contributes to model convergence. We perform experiments on two video grounding benchmarks of ActivityNet Captions and TACoS, demonstrating that our PRVG can significantly outperform previous methods. We also perform in-depth studies to investigate the effectiveness of parallel regression paradigm on video grounding.
Zelu Deng, Yujie Zhong, Sheng Guo, Weilin Huang
This work aims at improving instance retrieval with self-supervision. We find that fine-tuning using the recently developed self-supervised (SSL) learning methods, such as SimCLR and MoCo, fails to improve the performance of instance retrieval. In this work, we identify that the learnt representations for instance retrieval should be invariant to large variations in viewpoint and background etc., whereas self-augmented positives applied by the current SSL methods can not provide strong enough signals for learning robust instance-level representations. To overcome this problem, we propose InsCLR, a new SSL method that builds on the \textit{instance-level} contrast, to learn the intra-class invariance by dynamically mining meaningful pseudo positive samples from both mini-batches and a memory bank during training. Extensive experiments demonstrate that InsCLR achieves similar or even better performance than the state-of-the-art SSL methods on instance retrieval. Code is available at https://github.com/zeludeng/insclr.
Shiwen Zhang, Sheng Guo, Limin Wang, Weilin Huang, Matthew R. Scott
In this work, we propose Knowledge Integration Networks (referred as KINet) for video action recognition. KINet is capable of aggregating meaningful context features which are of great importance to identifying an action, such as human information and scene context. We design a three-branch architecture consisting of a main branch for action recognition, and two auxiliary branches for human parsing and scene recognition which allow the model to encode the knowledge of human and scene for action recognition. We explore two pre-trained models as teacher networks to distill the knowledge of human and scene for training the auxiliary tasks of KINet. Furthermore, we propose a two-level knowledge encoding mechanism which contains a Cross Branch Integration (CBI) module for encoding the auxiliary knowledge into medium-level convolutional features, and an Action Knowledge Graph (AKG) for effectively fusing high-level context information. This results in an end-to-end trainable framework where the three tasks can be trained collaboratively, allowing the model to compute strong context knowledge efficiently. The proposed KINet achieves the state-of-the-art performance on a large-scale action recognition benchmark Kinetics-400, with a top-1 accuracy of 77.8%. We further demonstrate that our KINet has strong capability by transferring the Kinetics-trained model to UCF-101, where it obtains 97.8% top-1 accuracy.
Haozhi Zhang, Xun Wang, Weilin Huang, Matthew R. Scott
Pair-wise loss functions have been extensively studied and shown to continuously improve the performance of deep metric learning (DML). However, they are primarily designed with intuition based on simple toy examples, and experimentally identifying the truly effective design is difficult in complicated, real-world cases. In this paper, we provide a new methodology for systematically studying weighting strategies of various pair-wise loss functions, and rethink pair weighting with an embedding memory. We delve into the weighting mechanisms by decomposing the pair-wise functions, and study positive and negative weights separately using direct weight assignment. This allows us to study various weighting functions deeply and systematically via weight curves, and identify a number of meaningful, comprehensive and insightful facts, which come up with our key observation on memory-based DML: it is critical to mine hard negatives and discard easy negatives which are less informative and redundant, but weighting on positive pairs is not helpful. This results in an efficient but surprisingly simple rule to design the weighting scheme, making it significantly different from existing mini-batch based methods which design various sophisticated loss functions to weight pairs carefully. Finally, we conduct extensive experiments on three large-scale visual retrieval benchmarks, and demonstrate the superiority of memory-based DML over recent mini-batch based approaches, by using a simple contrastive loss with momentum-updated memory.
Tong He, Zhi Tian, Weilin Huang, Chunhua Shen, Yu Qiao, Changming Sun
Text detection and recognition in natural images have long been considered as two separate tasks that are processed sequentially. Training of two tasks in a unified framework is non-trivial due to significant dif- ferences in optimisation difficulties. In this work, we present a conceptually simple yet efficient framework that simultaneously processes the two tasks in one shot. Our main contributions are three-fold: 1) we propose a novel text-alignment layer that allows it to precisely compute convolutional features of a text instance in ar- bitrary orientation, which is the key to boost the per- formance; 2) a character attention mechanism is introduced by using character spatial information as explicit supervision, leading to large improvements in recognition; 3) two technologies, together with a new RNN branch for word recognition, are integrated seamlessly into a single model which is end-to-end trainable. This allows the two tasks to work collaboratively by shar- ing convolutional features, which is critical to identify challenging text instances. Our model achieves impressive results in end-to-end recognition on the ICDAR2015 dataset, significantly advancing most recent results, with improvements of F-measure from (0.54, 0.51, 0.47) to (0.82, 0.77, 0.63), by using a strong, weak and generic lexicon respectively. Thanks to joint training, our method can also serve as a good detec- tor by achieving a new state-of-the-art detection performance on two datasets.
Junke Wang, Zhi Tian, Xun Wang, Xinyu Zhang, Weilin Huang, Zuxuan Wu, Yu-Gang Jiang
This work presents SimpleAR, a vanilla autoregressive visual generation framework without complex architecure modifications. Through careful exploration of training and inference optimization, we demonstrate that: 1) with only 0.5B parameters, our model can generate 1024x1024 resolution images with high fidelity, and achieve competitive results on challenging text-to-image benchmarks, e.g., 0.59 on GenEval and 79.66 on DPG; 2) both supervised fine-tuning (SFT) and Group Relative Policy Optimization (GRPO) training could lead to significant improvements on generation aesthectics and prompt alignment; and 3) when optimized with inference acceleraton techniques like vLLM, the time for SimpleAR to generate an 1024x1024 image could be reduced to around 14 seconds. By sharing these findings and open-sourcing the code, we hope to reveal the potential of autoregressive visual generation and encourage more participation in this research field. Code is available at https://github.com/wdrink/SimpleAR.
Team Seedance, Heyi Chen, Siyan Chen, Xin Chen, Yanfei Chen, Ying Chen, Zhuo Chen, Feng Cheng, Tianheng Cheng, Xinqi Cheng, Xuyan Chi, Jian Cong, Jing Cui, Qinpeng Cui, Qide Dong, Junliang Fan, Jing Fang, Zetao Fang, Chengjian Feng, Han Feng, Mingyuan Gao, Yu Gao, Dong Guo, Qiushan Guo, Boyang Hao, Qingkai Hao, Bibo He, Qian He, Tuyen Hoang, Ruoqing Hu, Xi Hu, Weilin Huang, Zhaoyang Huang, Zhongyi Huang, Donglei Ji, Siqi Jiang, Wei Jiang, Yunpu Jiang, Zhuo Jiang, Ashley Kim, Jianan Kong, Zhichao Lai, Shanshan Lao, Yichong Leng, Ai Li, Feiya Li, Gen Li, Huixia Li, JiaShi Li, Liang Li, Ming Li, Shanshan Li, Tao Li, Xian Li, Xiaojie Li, Xiaoyang Li, Xingxing Li, Yameng Li, Yifu Li, Yiying Li, Chao Liang, Han Liang, Jianzhong Liang, Ying Liang, Zhiqiang Liang, Wang Liao, Yalin Liao, Heng Lin, Kengyu Lin, Shanchuan Lin, Xi Lin, Zhijie Lin, Feng Ling, Fangfang Liu, Gaohong Liu, Jiawei Liu, Jie Liu, Jihao Liu, Shouda Liu, Shu Liu, Sichao Liu, Songwei Liu, Xin Liu, Xue Liu, Yibo Liu, Zikun Liu, Zuxi Liu, Junlin Lyu, Lecheng Lyu, Qian Lyu, Han Mu, Xiaonan Nie, Jingzhe Ning, Xitong Pan, Yanghua Peng, Lianke Qin, Xueqiong Qu, Yuxi Ren, Kai Shen, Guang Shi, Lei Shi, Yan Song, Yinglong Song, Fan Sun, Li Sun, Renfei Sun, Yan Sun, Zeyu Sun, Wenjing Tang, Yaxue Tang, Zirui Tao, Feng Wang, Furui Wang, Jinran Wang, Junkai Wang, Ke Wang, Kexin Wang, Qingyi Wang, Rui Wang, Sen Wang, Shuai Wang, Tingru Wang, Weichen Wang, Xin Wang, Yanhui Wang, Yue Wang, Yuping Wang, Yuxuan Wang, Ziyu Wang, Guoqiang Wei, Wanru Wei, Di Wu, Guohong Wu, Hanjie Wu, Jian Wu, Jie Wu, Ruolan Wu, Xinglong Wu, Yonghui Wu, Ruiqi Xia, Liang Xiang, Fei Xiao, XueFeng Xiao, Pan Xie, Shuangyi Xie, Shuang Xu, Jinlan Xue, Shen Yan, Bangbang Yang, Ceyuan Yang, Jiaqi Yang, Runkai Yang, Tao Yang, Yang Yang, Yihang Yang, ZhiXian Yang, Ziyan Yang, Songting Yao, Yifan Yao, Zilyu Ye, Bowen Yu, Jian Yu, Chujie Yuan, Linxiao Yuan, Sichun Zeng, Weihong Zeng, Xuejiao Zeng, Yan Zeng, Chuntao Zhang, Heng Zhang, Jingjie Zhang, Kuo Zhang, Liang Zhang, Liying Zhang, Manlin Zhang, Ting Zhang, Weida Zhang, Xiaohe Zhang, Xinyan Zhang, Yan Zhang, Yuan Zhang, Zixiang Zhang, Fengxuan Zhao, Huating Zhao, Yang Zhao, Hao Zheng, Jianbin Zheng, Xiaozheng Zheng, Yangyang Zheng, Yijie Zheng, Jiexin Zhou, Jiahui Zhu, Kuan Zhu, Shenhan Zhu, Wenjia Zhu, Benhui Zou, Feilong Zuo
Jie Liu, Zilyu Ye, Linxiao Yuan, Shenhan Zhu, Yu Gao, Jie Wu, Kunchang Li, Xionghui Wang, Xiaonan Nie, Weilin Huang, Wanli Ouyang
Unified models capable of interleaved generation have emerged as a promising paradigm, with the community increasingly converging on autoregressive modeling for text and flow matching for image generation. To advance this direction, we propose a unified reinforcement learning framework tailored for interleaved generation. We validate our approach on its fundamental unit: a single round of reasoning-driven image generation, where the model first expands the user prompt through reasoning, followed by image synthesis. Formulating this multimodal generation process as a Markov Decision Process with sparse terminal rewards, we introduce UniGRPO to jointly optimize text and image generation policies using GRPO. Adopting a minimalist methodology to avoid over-design, we leverage established training recipes for both modalities by seamlessly integrating standard GRPO for reasoning and FlowGRPO for visual synthesis. To ensure scalability to multi-round interleaved generation, we introduce two critical modifications to the original FlowGRPO: (1) eliminating classifier-free guidance to maintain linear, unbranched rollouts, which is essential for scaling to complex scenarios involving multi-turn interactions and multi-condition generation (e.g., editing); and (2) replacing the standard latent KL penalty with an MSE penalty directly on the velocity fields, providing a more robust and direct regularization signal to mitigate reward hacking effectively. Our experiments demonstrate that this unified training recipe significantly enhances image generation quality through reasoning, providing a robust and scalable baseline for the future post-training of fully interleaved models.
Weifeng Ge, Sheng Guo, Weilin Huang, Matthew R. Scott
Weakly-supervised instance segmentation aims to detect and segment object instances precisely, given imagelevel labels only. Unlike previous methods which are composed of multiple offline stages, we propose Sequential Label Propagation and Enhancement Networks (referred as Label-PEnet) that progressively transform image-level labels to pixel-wise labels in a coarse-to-fine manner. We design four cascaded modules including multi-label classification, object detection, instance refinement and instance segmentation, which are implemented sequentially by sharing the same backbone. The cascaded pipeline is trained alternatively with a curriculum learning strategy that generalizes labels from high-level images to low-level pixels gradually with increasing accuracy. In addition, we design a proposal calibration module to explore the ability of classification networks to find key pixels that identify object parts, which serves as a post validation strategy running in the inverse order. We evaluate the efficiency of our Label-PEnet in mining instance masks on standard benchmarks: PASCAL VOC 2007 and 2012. Experimental results show that Label-PEnet outperforms the state-of-the-art algorithms by a clear margin, and obtains comparable performance even with the fully-supervised approaches.
Yujie Zhong, Zelu Deng, Sheng Guo, Matthew R. Scott, Weilin Huang
Region Proposal Network (RPN) provides strong support for handling the scale variation of objects in two-stage object detection. For one-stage detectors which do not have RPN, it is more demanding to have powerful sub-networks capable of directly capturing objects of unknown sizes. To enhance such capability, we propose an extremely efficient neural architecture search method, named Fast And Diverse (FAD), to better explore the optimal configuration of receptive fields and convolution types in the sub-networks for one-stage detectors. FAD consists of a designed search space and an efficient architecture search algorithm. The search space contains a rich set of diverse transformations designed specifically for object detection. To cope with the designed search space, a novel search algorithm termed Representation Sharing (RepShare) is proposed to effectively identify the best combinations of the defined transformations. In our experiments, FAD obtains prominent improvements on two types of one-stage detectors with various backbones. In particular, our FAD detector achieves 46.4 AP on MS-COCO (under single-scale testing), outperforming the state-of-the-art detectors, including the most recent NAS-based detectors, Auto-FPN (searched for 16 GPU-days) and NAS-FCOS (28 GPU-days), while significantly reduces the search cost to 0.6 GPU-days. Beyond object detection, we further demonstrate the generality of FAD on the more challenging instance segmentation, and expect it to benefit more tasks.
Haoxian Tan, Sheng Guo, Yujie Zhong, Matthew R. Scott, Weilin Huang
Differentiable architecture search is prevalent in the field of NAS because of its simplicity and efficiency, where two paradigms, multi-path algorithms and single-path methods, are dominated. Multi-path framework (e.g. DARTS) is intuitive but suffers from memory usage and training collapse. Single-path methods (e.g.GDAS and ProxylessNAS) mitigate the memory issue and shrink the gap between searching and evaluation but sacrifice the performance. In this paper, we propose a conceptually simple yet efficient method to bridge these two paradigms, referred as Mutually-aware Sub-Graphs Differentiable Architecture Search (MSG-DAS). The core of our framework is a differentiable Gumbel-TopK sampler that produces multiple mutually exclusive single-path sub-graphs. To alleviate the severer skip-connect issue brought by multiple sub-graphs setting, we propose a Dropblock-Identity module to stabilize the optimization. To make best use of the available models (super-net and sub-graphs), we introduce a memory-efficient super-net guidance distillation to improve training. The proposed framework strikes a balance between flexible memory usage and searching quality. We demonstrate the effectiveness of our methods on ImageNet and CIFAR10, where the searched models show a comparable performance as the most recent approaches.
Chengjian Feng, Yujie Zhong, Weilin Huang
The conventional detectors tend to make imbalanced classification and suffer performance drop, when the distribution of the training data is severely skewed. In this paper, we propose to use the mean classification score to indicate the classification accuracy for each category during training. Based on this indicator, we balance the classification via an Equilibrium Loss (EBL) and a Memory-augmented Feature Sampling (MFS) method. Specifically, EBL increases the intensity of the adjustment of the decision boundary for the weak classes by a designed score-guided loss margin between any two classes. On the other hand, MFS improves the frequency and accuracy of the adjustment of the decision boundary for the weak classes through over-sampling the instance features of those classes. Therefore, EBL and MFS work collaboratively for finding the classification equilibrium in long-tailed detection, and dramatically improve the performance of tail classes while maintaining or even improving the performance of head classes. We conduct experiments on LVIS using Mask R-CNN with various backbones including ResNet-50-FPN and ResNet-101-FPN to show the superiority of the proposed method. It improves the detection performance of tail classes by 15.6 AP, and outperforms the most recent long-tailed object detectors by more than 1 AP. Code is available at https://github.com/fcjian/LOCE.
Chengjian Feng, Yujie Zhong, Yu Gao, Matthew R. Scott, Weilin Huang
One-stage object detection is commonly implemented by optimizing two sub-tasks: object classification and localization, using heads with two parallel branches, which might lead to a certain level of spatial misalignment in predictions between the two tasks. In this work, we propose a Task-aligned One-stage Object Detection (TOOD) that explicitly aligns the two tasks in a learning-based manner. First, we design a novel Task-aligned Head (T-Head) which offers a better balance between learning task-interactive and task-specific features, as well as a greater flexibility to learn the alignment via a task-aligned predictor. Second, we propose Task Alignment Learning (TAL) to explicitly pull closer (or even unify) the optimal anchors for the two tasks during training via a designed sample assignment scheme and a task-aligned loss. Extensive experiments are conducted on MS-COCO, where TOOD achieves a 51.1 AP at single-model single-scale testing. This surpasses the recent one-stage detectors by a large margin, such as ATSS (47.7 AP), GFL (48.2 AP), and PAA (49.0 AP), with fewer parameters and FLOPs. Qualitative results also demonstrate the effectiveness of TOOD for better aligning the tasks of object classification and localization. Code is available at https://github.com/fcjian/TOOD.
Pan He, Weilin Huang, Tong He, Qile Zhu, Yu Qiao, Xiaolin Li
We present a novel single-shot text detector that directly outputs word-level bounding boxes in a natural image. We propose an attention mechanism which roughly identifies text regions via an automatically learned attentional map. This substantially suppresses background interference in the convolutional features, which is the key to producing accurate inference of words, particularly at extremely small sizes. This results in a single model that essentially works in a coarse-to-fine manner. It departs from recent FCN- based text detectors which cascade multiple FCN models to achieve an accurate prediction. Furthermore, we develop a hierarchical inception module which efficiently aggregates multi-scale inception features. This enhances local details, and also encodes strong context information, allow- ing the detector to work reliably on multi-scale and multi- orientation text with single-scale images. Our text detector achieves an F-measure of 77% on the ICDAR 2015 bench- mark, advancing the state-of-the-art results in [18, 28]. Demo is available at: http://sstd.whuang.org/.
Spyridon Bakas, Mauricio Reyes, Andras Jakab, Stefan Bauer, Markus Rempfler, Alessandro Crimi, Russell Takeshi Shinohara, Christoph Berger, Sung Min Ha, Martin Rozycki, Marcel Prastawa, Esther Alberts, Jana Lipkova, John Freymann, Justin Kirby, Michel Bilello, Hassan Fathallah-Shaykh, Roland Wiest, Jan Kirschke, Benedikt Wiestler, Rivka Colen, Aikaterini Kotrotsou, Pamela Lamontagne, Daniel Marcus, Mikhail Milchenko, Arash Nazeri, Marc-Andre Weber, Abhishek Mahajan, Ujjwal Baid, Elizabeth Gerstner, Dongjin Kwon, Gagan Acharya, Manu Agarwal, Mahbubul Alam, Alberto Albiol, Antonio Albiol, Francisco J. Albiol, Varghese Alex, Nigel Allinson, Pedro H. A. Amorim, Abhijit Amrutkar, Ganesh Anand, Simon Andermatt, Tal Arbel, Pablo Arbelaez, Aaron Avery, Muneeza Azmat, Pranjal B., W Bai, Subhashis Banerjee, Bill Barth, Thomas Batchelder, Kayhan Batmanghelich, Enzo Battistella, Andrew Beers, Mikhail Belyaev, Martin Bendszus, Eze Benson, Jose Bernal, Halandur Nagaraja Bharath, George Biros, Sotirios Bisdas, James Brown, Mariano Cabezas, Shilei Cao, Jorge M. Cardoso, Eric N Carver, Adrià Casamitjana, Laura Silvana Castillo, Marcel Catà, Philippe Cattin, Albert Cerigues, Vinicius S. Chagas, Siddhartha Chandra, Yi-Ju Chang, Shiyu Chang, Ken Chang, Joseph Chazalon, Shengcong Chen, Wei Chen, Jefferson W Chen, Zhaolin Chen, Kun Cheng, Ahana Roy Choudhury, Roger Chylla, Albert Clérigues, Steven Colleman, Ramiro German Rodriguez Colmeiro, Marc Combalia, Anthony Costa, Xiaomeng Cui, Zhenzhen Dai, Lutao Dai, Laura Alexandra Daza, Eric Deutsch, Changxing Ding, Chao Dong, Shidu Dong, Wojciech Dudzik, Zach Eaton-Rosen, Gary Egan, Guilherme Escudero, Théo Estienne, Richard Everson, Jonathan Fabrizio, Yong Fan, Longwei Fang, Xue Feng, Enzo Ferrante, Lucas Fidon, Martin Fischer, Andrew P. French, Naomi Fridman, Huan Fu, David Fuentes, Yaozong Gao, Evan Gates, David Gering, Amir Gholami, Willi Gierke, Ben Glocker, Mingming Gong, Sandra González-Villá, T. Grosges, Yuanfang Guan, Sheng Guo, Sudeep Gupta, Woo-Sup Han, Il Song Han, Konstantin Harmuth, Huiguang He, Aura Hernández-Sabaté, Evelyn Herrmann, Naveen Himthani, Winston Hsu, Cheyu Hsu, Xiaojun Hu, Xiaobin Hu, Yan Hu, Yifan Hu, Rui Hua, Teng-Yi Huang, Weilin Huang, Sabine Van Huffel, Quan Huo, Vivek HV, Khan M. Iftekharuddin, Fabian Isensee, Mobarakol Islam, Aaron S. Jackson, Sachin R. Jambawalikar, Andrew Jesson, Weijian Jian, Peter Jin, V Jeya Maria Jose, Alain Jungo, B Kainz, Konstantinos Kamnitsas, Po-Yu Kao, Ayush Karnawat, Thomas Kellermeier, Adel Kermi, Kurt Keutzer, Mohamed Tarek Khadir, Mahendra Khened, Philipp Kickingereder, Geena Kim, Nik King, Haley Knapp, Urspeter Knecht, Lisa Kohli, Deren Kong, Xiangmao Kong, Simon Koppers, Avinash Kori, Ganapathy Krishnamurthi, Egor Krivov, Piyush Kumar, Kaisar Kushibar, Dmitrii Lachinov, Tryphon Lambrou, Joon Lee, Chengen Lee, Yuehchou Lee, M Lee, Szidonia Lefkovits, Laszlo Lefkovits, James Levitt, Tengfei Li, Hongwei Li, Wenqi Li, Hongyang Li, Xiaochuan Li, Yuexiang Li, Heng Li, Zhenye Li, Xiaoyu Li, Zeju Li, XiaoGang Li, Wenqi Li, Zheng-Shen Lin, Fengming Lin, Pietro Lio, Chang Liu, Boqiang Liu, Xiang Liu, Mingyuan Liu, Ju Liu, Luyan Liu, Xavier Llado, Marc Moreno Lopez, Pablo Ribalta Lorenzo, Zhentai Lu, Lin Luo, Zhigang Luo, Jun Ma, Kai Ma, Thomas Mackie, Anant Madabushi, Issam Mahmoudi, Klaus H. Maier-Hein, Pradipta Maji, CP Mammen, Andreas Mang, B. S. Manjunath, Michal Marcinkiewicz, S McDonagh, Stephen McKenna, Richard McKinley, Miriam Mehl, Sachin Mehta, Raghav Mehta, Raphael Meier, Christoph Meinel, Dorit Merhof, Craig Meyer, Robert Miller, Sushmita Mitra, Aliasgar Moiyadi, David Molina-Garcia, Miguel A. B. Monteiro, Grzegorz Mrukwa, Andriy Myronenko, Jakub Nalepa, Thuyen Ngo, Dong Nie, Holly Ning, Chen Niu, Nicholas K Nuechterlein, Eric Oermann, Arlindo Oliveira, Diego D. C. Oliveira, Arnau Oliver, Alexander F. I. Osman, Yu-Nian Ou, Sebastien Ourselin, Nikos Paragios, Moo Sung Park, Brad Paschke, J. Gregory Pauloski, Kamlesh Pawar, Nick Pawlowski, Linmin Pei, Suting Peng, Silvio M. Pereira, Julian Perez-Beteta, Victor M. Perez-Garcia, Simon Pezold, Bao Pham, Ashish Phophalia, Gemma Piella, G. N. Pillai, Marie Piraud, Maxim Pisov, Anmol Popli, Michael P. Pound, Reza Pourreza, Prateek Prasanna, Vesna Prkovska, Tony P. Pridmore, Santi Puch, Élodie Puybareau, Buyue Qian, Xu Qiao, Martin Rajchl, Swapnil Rane, Michael Rebsamen, Hongliang Ren, Xuhua Ren, Karthik Revanuru, Mina Rezaei, Oliver Rippel, Luis Carlos Rivera, Charlotte Robert, Bruce Rosen, Daniel Rueckert, Mohammed Safwan, Mostafa Salem, Joaquim Salvi, Irina Sanchez, Irina Sánchez, Heitor M. Santos, Emmett Sartor, Dawid Schellingerhout, Klaudius Scheufele, Matthew R. Scott, Artur A. Scussel, Sara Sedlar, Juan Pablo Serrano-Rubio, N. Jon Shah, Nameetha Shah, Mazhar Shaikh, B. Uma Shankar, Zeina Shboul, Haipeng Shen, Dinggang Shen, Linlin Shen, Haocheng Shen, Varun Shenoy, Feng Shi, Hyung Eun Shin, Hai Shu, Diana Sima, M Sinclair, Orjan Smedby, James M. Snyder, Mohammadreza Soltaninejad, Guidong Song, Mehul Soni, Jean Stawiaski, Shashank Subramanian, Li Sun, Roger Sun, Jiawei Sun, Kay Sun, Yu Sun, Guoxia Sun, Shuang Sun, Yannick R Suter, Laszlo Szilagyi, Sanjay Talbar, Dacheng Tao, Dacheng Tao, Zhongzhao Teng, Siddhesh Thakur, Meenakshi H Thakur, Sameer Tharakan, Pallavi Tiwari, Guillaume Tochon, Tuan Tran, Yuhsiang M. Tsai, Kuan-Lun Tseng, Tran Anh Tuan, Vadim Turlapov, Nicholas Tustison, Maria Vakalopoulou, Sergi Valverde, Rami Vanguri, Evgeny Vasiliev, Jonathan Ventura, Luis Vera, Tom Vercauteren, C. A. Verrastro, Lasitha Vidyaratne, Veronica Vilaplana, Ajeet Vivekanandan, Guotai Wang, Qian Wang, Chiatse J. Wang, Weichung Wang, Duo Wang, Ruixuan Wang, Yuanyuan Wang, Chunliang Wang, Guotai Wang, Ning Wen, Xin Wen, Leon Weninger, Wolfgang Wick, Shaocheng Wu, Qiang Wu, Yihong Wu, Yong Xia, Yanwu Xu, Xiaowen Xu, Peiyuan Xu, Tsai-Ling Yang, Xiaoping Yang, Hao-Yu Yang, Junlin Yang, Haojin Yang, Guang Yang, Hongdou Yao, Xujiong Ye, Changchang Yin, Brett Young-Moxon, Jinhua Yu, Xiangyu Yue, Songtao Zhang, Angela Zhang, Kun Zhang, Xuejie Zhang, Lichi Zhang, Xiaoyue Zhang, Yazhuo Zhang, Lei Zhang, Jianguo Zhang, Xiang Zhang, Tianhao Zhang, Sicheng Zhao, Yu Zhao, Xiaomei Zhao, Liang Zhao, Yefeng Zheng, Liming Zhong, Chenhong Zhou, Xiaobing Zhou, Fan Zhou, Hongtu Zhu, Jin Zhu, Ying Zhuge, Weiwei Zong, Jayashree Kalpathy-Cramer, Keyvan Farahani, Christos Davatzikos, Koen van Leemput, Bjoern Menze