分享

In this paper we study the application of convolutional neural networks for jointly detecting objects depicted in still images and estimating their 3D pose. We identify different feature representations of oriented objects, and energies that lead a network to learn this representations. The choice of the representation is crucial since the pose of an object has a natural, continuous structure while its category is a discrete variable. We evaluate the different approaches on the joint object detection and pose estimation task of the Pascal3D+ benchmark using Average Viewpoint Precision. We show that a classification approach on discretized viewpoints achieves state-of-the-art performance for joint object detection and pose estimation, and significantly outperforms existing baselines on this benchmark.

We train a model to generate images from multimodal prompts of interleaved text and images such as "a <picture of a man> man and his <picture of a dog> dog in an <picture of a cartoon> animated style." We bootstrap a multimodal dataset by extracting semantically meaningful image crops corresponding to words in the image captions of synthetically generated and publicly available text-image data. Our model, MUMU, is composed of a vision-language model encoder with a diffusion decoder and is trained on a single 8xH100 GPU node. Despite being only trained on crops from the same image, MUMU learns to compose inputs from different images into a coherent output. For example, an input of a realistic person and a cartoon will output the same person in the cartoon style, and an input of a standing subject and a scooter will output the subject riding the scooter. As a result, our model generalizes to tasks such as style transfer and character consistency. Our results show the promise of using multimodal models as general purpose controllers for image generation.

Program refinement involves correctness-preserving transformations from formal high-level specification statements into executable programs. Traditional verification tool support for program refinement is highly interactive and lacks automation. On the other hand, the emergence of large language models (LLMs) enables automatic code generations from informal natural language specifications. However, code generated by LLMs is often unreliable. Moreover, the opaque procedure from specification to code provided by LLM is an uncontrolled black box. We propose LLM4PR, a tool that combines formal program refinement techniques with informal LLM-based methods to (1) transform the specification to preconditions and postconditions, (2) automatically build prompts based on refinement calculus, (3) interact with LLM to generate code, and finally, (4) verify that the generated code satisfies the conditions of refinement calculus, thus guaranteeing the correctness of the code. We have implemented our tool using GPT4, Coq, and Coqhammer, and evaluated it on the HumanEval and EvalPlus datasets.

Human modelling and pose estimation stands at the crossroads of Computer Vision, Computer Graphics, and Machine Learning. This paper presents a thorough investigation of this interdisciplinary field, examining various algorithms, methodologies, and practical applications. It explores the diverse range of sensor technologies relevant to this domain and delves into a wide array of application areas. Additionally, we discuss the challenges and advancements in 2D and 3D human modelling methodologies, along with popular datasets, metrics, and future research directions. The main contribution of this paper lies in its up-to-date comparison of state-of-the-art (SOTA) human pose estimation algorithms in both 2D and 3D domains. By providing this comprehensive overview, the paper aims to enhance understanding of 3D human modelling and pose estimation, offering insights into current SOTA achievements, challenges, and future prospects within the field.

This paper proposes a novel zero-shot composed image retrieval (CIR) method considering the query-target relationship by masked image-text pairs. The objective of CIR is to retrieve the target image using a query image and a query text. Existing methods use a textual inversion network to convert the query image into a pseudo word to compose the image and text and use a pre-trained visual-language model to realize the retrieval. However, they do not consider the query-target relationship to train the textual inversion network to acquire information for retrieval. In this paper, we propose a novel zero-shot CIR method that is trained end-to-end using masked image-text pairs. By exploiting the abundant image-text pairs that are convenient to obtain with a masking strategy for learning the query-target relationship, it is expected that accurate zero-shot CIR using a retrieval-focused textual inversion network can be realized. Experimental results show the effectiveness of the proposed method.

This paper presents the development of a speech synthesis system for the LIMMITS'24 Challenge, focusing primarily on Track 2. The objective of the challenge is to establish a multi-speaker, multi-lingual Indic Text-to-Speech system with voice cloning capabilities, covering seven Indian languages with both male and female speakers. The system was trained using challenge data and fine-tuned for few-shot voice cloning on target speakers. Evaluation included both mono-lingual and cross-lingual synthesis across all seven languages, with subjective tests assessing naturalness and speaker similarity. Our system uses the VITS2 architecture, augmented with a multi-lingual ID and a BERT model to enhance contextual language comprehension. In Track 1, where no additional data usage was permitted, our model achieved a Speaker Similarity score of 4.02. In Track 2, which allowed the use of extra data, it attained a Speaker Similarity score of 4.17.

Forced alignment (FA) plays a key role in speech research through the automatic time alignment of speech signals with corresponding text transcriptions. Despite the move towards end-to-end architectures for speech technology, FA is still dominantly achieved through a classic GMM-HMM acoustic model. This work directly compares alignment performance from leading automatic speech recognition (ASR) methods, WhisperX and Massively Multilingual Speech Recognition (MMS), against a Kaldi-based GMM-HMM system, the Montreal Forced Aligner (MFA). Performance was assessed on the manually aligned TIMIT and Buckeye datasets, with comparisons conducted only on words correctly recognized by WhisperX and MMS. The MFA outperformed both WhisperX and MMS, revealing a shortcoming of modern ASR systems. These findings highlight the need for advancements in forced alignment and emphasize the importance of integrating traditional expertise with modern innovation to foster progress. Index Terms: forced alignment, phoneme alignment, word alignment

Photo-Sharing Multi-modal dialogue generation requires a dialogue agent not only to generate text responses but also to share photos at the proper moment. Using image text caption as the bridge, a pipeline model integrates an image caption model, a text generation model, and an image generation model to handle this complex multi-modal task. However, representing the images with text captions may loss important visual details and information and cause error propagation in the complex dialogue system. Besides, the pipeline model isolates the three models separately because discrete image text captions hinder end-to-end gradient propagation. We propose the first end-to-end model for photo-sharing multi-modal dialogue generation, which integrates an image perceptron and an image generator with a large language model. The large language model employs the Q-Former to perceive visual images in the input end. For image generation in the output end, we propose a dynamic vocabulary transformation matrix and use straight-through and gumbel-softmax techniques to align the large language model and stable diffusion model and achieve end-to-end gradient propagation. We perform experiments on PhotoChat and DialogCC datasets to evaluate our end-to-end model. Compared with pipeline models, the end-to-end model gains state-of-the-art performances on various metrics of text and image generation. More analysis experiments also verify the effectiveness of the end-to-end model for photo-sharing multi-modal dialogue generation.

Open-world 3D generation has recently attracted considerable attention. While many single-image-to-3D methods have yielded visually appealing outcomes, they often lack sufficient controllability and tend to produce hallucinated regions that may not align with users' expectations. In this paper, we explore an important scenario in which the input consists of one or a few unposed 2D images of a single object, with little or no overlap. We propose a novel method, SpaRP, to reconstruct a 3D textured mesh and estimate the relative camera poses for these sparse-view images. SpaRP distills knowledge from 2D diffusion models and finetunes them to implicitly deduce the 3D spatial relationships between the sparse views. The diffusion model is trained to jointly predict surrogate representations for camera poses and multi-view images of the object under known poses, integrating all information from the input sparse views. These predictions are then leveraged to accomplish 3D reconstruction and pose estimation, and the reconstructed 3D model can be used to further refine the camera poses of input views. Through extensive experiments on three datasets, we demonstrate that our method not only significantly outperforms baseline methods in terms of 3D reconstruction quality and pose prediction accuracy but also exhibits strong efficiency. It requires only about 20 seconds to produce a textured mesh and camera poses for the input views. Project page: https://chaoxu.xyz/sparp.

Recent work in image and video generation has been adopting the autoregressive LLM architecture due to its generality and potentially easy integration into multi-modal systems. The crux of applying autoregressive training in language generation to visual generation is discretization -- representing continuous data like images and videos as discrete tokens. Common methods of discretizing images and videos include modeling raw pixel values, which are prohibitively lengthy, or vector quantization, which requires convoluted pre-hoc training. In this work, we propose to directly model images and videos as compressed files saved on computers via canonical codecs (e.g., JPEG, AVC/H.264). Using the default Llama architecture without any vision-specific modifications, we pretrain JPEG-LM from scratch to generate images (and AVC-LM to generate videos as a proof of concept), by directly outputting compressed file bytes in JPEG and AVC formats. Evaluation of image generation shows that this simple and straightforward approach is more effective than pixel-based modeling and sophisticated vector quantization baselines (on which our method yields a 31% reduction in FID). Our analysis shows that JPEG-LM has an especial advantage over vector quantization models in generating long-tail visual elements. Overall, we show that using canonical codec representations can help lower the barriers between language generation and visual generation, facilitating future research on multi-modal language/image/video LLMs.

Self-ensembling techniques with diverse reasoning paths such as Self-Consistency have demonstrated remarkable performance gains in text generation with Large Language Models (LLMs). However, such techniques depend on the availability of an accurate answer extraction process to aggregate across multiple outputs. Moreover, they acquire higher inference cost, in comparison to Greedy Decoding, due to generation of relatively higher number of output tokens. Research has shown that the free form text outputs from Self-Consistency can be aggregated reliably using LLMs to produce the final output. Additionally, recent advancements in LLM inference have demonstrated that usage of diverse exemplars in prompts have the ability to induce diversity in the LLM outputs. Such proven techniques can be easily extended to self-ensembling based approaches to achieve enhanced results in text generation. In this paper, we introduce PEDAL (Prompts based on Exemplar Diversity Aggregated using LLMs), a hybrid self-ensembling approach, that combines the strengths of diverse exemplar based prompts and LLM based aggregation to achieve improvement in overall performance. On the publicly available SVAMP and ARC datasets, our experiments reveal that PEDAL can achieve better accuracy than Greedy Decoding based strategies with lower inference cost compared to Self Consistency based approaches.

Deep learning (DL) models based on the transformer architecture have revolutionized many DL applications such as large language models (LLMs), vision transformers, audio generation, and time series prediction. Much of this progress has been fueled by distributed training, yet distributed communication remains a substantial bottleneck to training progress. This paper examines the communication behavior of transformer models - that is, how different parallelism schemes used in multi-node/multi-GPU DL Training communicate data in the context of transformers. We use GPT-based language models as a case study of the transformer architecture due to their ubiquity. We validate the empirical results obtained from our communication logs using analytical models. At a high level, our analysis reveals a need to optimize small message point-to-point communication further, correlations between sequence length, per-GPU throughput, model size, and optimizations used, and where to potentially guide further optimizations in framework and HPC middleware design and optimization.