As humans, we can modify our assumptions about a scene by imagining alternative objects or concepts in our minds. For example, we can easily anticipate the implications of the sun being overcast by rain clouds (e.g., the street will get wet) and accordingly prepare for that. In this paper, we introduce a new task/dataset called Commonsense Reasoning for Counterfactual Scene Imagination (COSIM) which is designed to evaluate the ability of AI systems to reason about scene change imagination. In this task/dataset, models are given an image and an initial question-response pair about the image. Next, a counterfactual imagined scene change (in textual form) is applied, and the model has to predict the new response to the initial question based on this scene change. We collect 3.5K high-quality and challenging data instances, with each instance consisting of an image, a commonsense question with a response, a description of a counterfactual change, a new response to the question, and three distractor responses. Our dataset contains various complex scene change types (such as object addition/removal/state change, event description, environment change, etc.) that require models to imagine many different scenarios and reason about the changed scenes. We present a baseline model based on a vision-language Transformer (i.e., LXMERT) and ablation studies. Through human evaluation, we demonstrate a large human-model performance gap, suggesting room for promising future work on this challenging counterfactual, scene imagination task.
more »
« less
FIXMYPOSE: Pose Correctional Captioning and Retrieval
Interest in physical therapy and individual exercises such as yoga/dance has increased alongside the well-being trend, and people globally enjoy such exercises at home/office via video streaming platforms. However, such exercises are hard to follow without expert guidance. Even if experts can help, it is almost impossible to give personalized feedback to every trainee remotely. Thus, automated pose correction systems are required more than ever, and we introduce a new captioning dataset named FixMyPose to address this need. We collect natural language descriptions of correcting a “current” pose to look like a “target” pose. To support a multilingual setup, we collect descriptions in both English and Hindi. The collected descriptions have interesting linguistic properties such as egocentric relations to the environment objects, analogous references, etc., requiring an understanding of spatial relations and commonsense knowledge about postures. Further, to avoid ML biases, we maintain a balance across characters with diverse demographics, who perform a variety of movements in several interior environments (e.g., homes, offices). From our FixMyPose dataset, we introduce two tasks: the pose-correctional-captioning task and its reverse, the target-pose-retrieval task. During the correctional-captioning task, models must generate the descriptions of how to move from the current to the target pose image, whereas in the retrieval task, models should select the correct target pose given the initial pose and the correctional description. We present strong cross-attention baseline models (uni/multimodal, RL, multilingual) and also show that our baselines are competitive with other models when evaluated on other image-difference datasets. We also propose new task-specific metrics (object-match, body-part-match, direction-match) and conduct human evaluation for more reliable evaluation, and we demonstrate a large human-model performance gap suggesting room for promising future work. Finally, to verify the sim-to-real transfer of our FixMyPose dataset, we collect a set of real images and show promising performance on these images. Data and code are available: https://fixmypose-unc.github.io.
more »
« less
- Award ID(s):
- 1840131
- NSF-PAR ID:
- 10301947
- Date Published:
- Journal Name:
- Proceedings of the AAAI Conference on Artificial Intelligence
- Volume:
- 35
- Issue:
- 14
- ISSN:
- 2374-3468
- Page Range / eLocation ID:
- 3910–3927
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Demand for image editing has been increasing as users' desire for expression is also increasing. However, for most users, image editing tools are not easy to use since the tools require certain expertise in photo effects and have complex interfaces. Hence, users might need someone to help edit their images, but having a personal dedicated human assistant for every user is impossible to scale. For that reason, an automated assistant system for image editing is desirable. Additionally, users want more image sources for diverse image editing works, and integrating an image search functionality into the editing tool is a potential remedy for this demand. Thus, we propose a dataset of an automated Conversational Agent for Image Search and Editing (CAISE). To our knowledge, this is the first dataset that provides conversational image search and editing annotations, where the agent holds a grounded conversation with users and helps them to search and edit images according to their requests. To build such a system, we first collect image search and editing conversations between pairs of annotators. The assistant-annotators are equipped with a customized image search and editing tool to address the requests from the user-annotators. The functions that the assistant-annotators conduct with the tool are recorded as executable commands, allowing the trained system to be useful for real-world application execution. We also introduce a generator-extractor baseline model for this task, which can adaptively select the source of the next token (i.e., from the vocabulary or from textual/visual contexts) for the executable command. This serves as a strong starting point while still leaving a large human-machine performance gap for useful future work.conversational image search and editing annotations, where the agent holds a grounded conversation with users and helps them to search and edit images according to their requests. To build such a system, we first collect image search and editing conversations between pairs of annotators. The assistant-annotators are equipped with a customized image search and editing tool to address the requests from the user-annotators. The functions that the assistant-annotators conduct with the tool are recorded as executable commands, allowing the trained system to be useful for real-world application execution. We also introduce a generator-extractor baseline model for this task, which can adaptively select the source of the next token (i.e., from the vocabulary or from textual/visual contexts) for the executable command. This serves as a strong starting point while still leaving a large human-machine performance gap for useful future work.more » « less
-
People are able to describe images using thousands of languages, but languages share only one visual world. The aim of this work is to use the learned intermediate visual representations from a deep convolutional neural network to transfer information across languages for which paired data is not available in any form. Our work proposes using backpropagation-based decoding coupled with transformer-based multilingual-multimodal language models in order to obtain translations between any languages used during training. We particularly show the capabilities of this approach in the translation of German-Japanese and Japanese-German sentence pairs, given a training data of images freely associated with text in English, German, and Japanese but for which no single image contains annotations in both Japanese and German. Moreover, we demonstrate that our approach is also generally useful in the multilingual image captioning task when sentences in a second language are available at test time. The results of our method also compare favorably in the Multi30k dataset against recently proposed methods that are also aiming to leverage images as an intermediate source of translations.more » « less
-
Telecystoscopy can lower the barrier to access critical urologic diagnostics for patients around the world. A major challenge for robotic control of flexible cystoscopes and intuitive teleoperation is the pose estimation of the scope tip. We propose a novel real-time camera localization method using video recordings from a prior cystoscopy and 3D bladder reconstruction to estimate cystoscope pose within the bladder during follow-up telecystoscopy. We map prior video frames into a low-dimensional space as a dictionary so that a new image can be likewise mapped to efficiently retrieve its nearest neighbor among the dictionary images. The cystoscope pose is then estimated by the correspondence among the new image, its nearest dictionary image, and the prior model from 3D reconstruction. We demonstrate performance of our methods using bladder phantoms with varying fidelity and a servo-controlled cystoscope to simulate the use case of bladder surveillance through telecystoscopy. The servo-controlled cystoscope with 3 degrees of freedom (angulation, roll, and insertion axes) was developed for collecting cystoscope videos from bladder phantoms. Cystoscope videos were acquired in a 2.5D bladder phantom (bladder-shape cross-section plus height) with a panorama of a urothelium attached to the inner surface. Scans of the 2.5D phantom were performed in separate arc trajectories each of which is generated by actuation on the angulation with a fixed roll and insertion length. We further included variance in moving speed, imaging distance and existence of bladder tumors. Cystoscope videos were also acquired in a water-filled 3D silicone bladder phantom with hand-painted vasculature. Scans of the 3D phantom were performed in separate circle trajectories each of which is generated by actuation on the roll axis under a fixed angulation and insertion length. These videos were used to create 3D reconstructions, dictionary sets, and test data sets for evaluating the computational efficiency and accuracy of our proposed method in comparison with a method based on global Scale-Invariant Feature Transform (SIFT) features, named SIFT-only. Our method can retrieve the nearest dictionary image for 94–100% of test frames in under 55[Formula: see text]ms per image, whereas the SIFT-only method can only find the image match for 56–100% of test frames in 6000–40000[Formula: see text]ms per image depending on size of the dictionary set and richness of SIFT features in the images. Our method, with a speed of around 20 Hz for the retrieval stage, is a promising tool for real-time image-based scope localization in robotic cystoscopy when prior cystoscopy images are available.more » « less
-
The prediction of human shifts of attention is a widely-studied question in both behavioral and computer vision, especially in the context of a free viewing task. However, search behavior, where the fixation scanpaths are highly dependent on the viewer’s goals, has received far less attention, even though visual search constitutes much of a person’s everyday behavior. One reason for this is the absence of real-world image datasets on which search models can be trained. In this paper we present a carefully created dataset for two target categories, microwaves and clocks, curated from the COCO2014 dataset. A total of 2183 images were presented to multiple participants, who were tasked to search for one of the two categories. This yields a total of 16184 validated fixations used for training, making our microwave-clock dataset currently one of the largest datasets of eye fixations in categorical search. We also present a 40-image testing dataset, where images depict both a microwave and a clock target. Distinct fixation patterns emerged depending on whether participants searched for a microwave (n=30) or a clock (n=30) in the same images, meaning that models need to predict different search scanpaths from the same pixel inputs. We report the results of several state-of-the-art deep network models that were trained and evaluated on these datasets. Collectively, these datasets and our protocol for evaluation provide what we hope will be a useful test-bed for the development of new methods for predicting category-specific visual search behavior.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
